U.S. patent application number 12/058120 was filed with the patent office on 2008-10-02 for method and system for managing a mortgage-backed securities index.
This patent application is currently assigned to CREDIT SUISSE SECURITIES (USA) LLC. Invention is credited to David RIOS, Matt RUPPEL.
Application Number | 20080243717 12/058120 |
Document ID | / |
Family ID | 25460294 |
Filed Date | 2008-10-02 |
United States Patent
Application |
20080243717 |
Kind Code |
A1 |
RIOS; David ; et
al. |
October 2, 2008 |
METHOD AND SYSTEM FOR MANAGING A MORTGAGE-BACKED SECURITIES
INDEX
Abstract
A system and method for generating and managing a generic
mortgage-backed securities index. Bonds are selected for the index
on a monthly basis. In order to determine which bonds will be
represented in the index during a particular month, a set of
calculations is performed during the second business week of the
preceding month. For the purpose of selecting bonds for the index,
all outstanding mortgage-backed securities are considered. They are
first aggregated into pools based on their coupon and original
term, and then their total outstanding principals are considered.
If the total principal outstanding of any 30-year coupon represents
more than a predetermined percentage such as 1.5% of the total,
then this 30-year coupon will be included in the Index. Similarly,
if the total principal outstanding on any 15-year coupon represents
more than a second predetermined percentage such as 0.4% of the
total, then this 15-year coupon will be included in the Index. The
performance of the Index is measured by its total return. An
algorithm for calculating the total return of the generic Index is
also provided. The total return of the index partially depends on
the relative weight assigned to each particular security included
in the index. The present invention provides a method of assigning
relative weights in accordance with relative proportions of
different individual securities in the index, and covers the
frequency of re-weighting.
Inventors: |
RIOS; David; (Long Island
City, NY) ; RUPPEL; Matt; (New York, NY) |
Correspondence
Address: |
OSTROLENK FABER GERB & SOFFEN
1180 AVENUE OF THE AMERICAS
NEW YORK
NY
100368403
US
|
Assignee: |
CREDIT SUISSE SECURITIES (USA)
LLC
New York
NY
|
Family ID: |
25460294 |
Appl. No.: |
12/058120 |
Filed: |
March 28, 2008 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
09931149 |
Aug 16, 2001 |
7353198 |
|
|
12058120 |
|
|
|
|
Current U.S.
Class: |
705/36R |
Current CPC
Class: |
G06Q 40/04 20130101;
G06Q 40/06 20130101; G06Q 40/025 20130101 |
Class at
Publication: |
705/36.R |
International
Class: |
G06Q 40/00 20060101
G06Q040/00; G06F 17/10 20060101 G06F017/10 |
Claims
1. A method for managing a mortgage-backed securities index,
comprising the steps of: selecting a set of mortgage-backed
securities to be included in said mortgage-backed securities index,
said set of mortgaged-backed securities being selected from all
outstanding mortgage-backed securities; assigning a relative weight
to each security within said selected set, said relative weight
being a relative proportion of total outstanding principal on said
each security to the total outstanding principal on all securities
within said selected set; calculating a total return of said
mortgage-backed securities index, said total return being based on
said assigned relative weight for said each security, and a total
return of said each security based on a same-day-settle price;
storing the calculated total return of said mortgage-backed
securities index in an index database; and outputting the
calculated total return of said mortgage-backed securities index to
a user, wherein said step of selecting a set of mortgage-backed
securities further comprises aggregating said all outstanding
mortgage-backed securities into a plurality of pools, each of said
pools comprising mortgage-backed securities having the same coupon
and the same original term, and determining inclusion criteria for
each pool within said plurality of pools, said inclusion criteria
being determined using an objective condition.
2. A method for managing a mortgage-backed securities index
according to claim 1 wherein said inclusion criteria is given by
the following equation: x c , t = [ a = { FNMA GNMA FHLMC } .rho. a
, c , t ] [ a = FNMA , GNMA , FHLMC c .di-elect cons. Z t = 180 ,
360 .rho. a , ct ] , ##EQU00018## wherein .rho..sub.a,c,t is the
total outstanding principal on said outstanding mortgage-backed
securities, a is an agency which issued said outstanding
mortgage-backed securities, c is a coupon value of said outstanding
mortgage-backed securities, and t is an original term of said
outstanding mortgage-backed securities.
3. A method for managing a mortgage-backed securities index
according to claim 1, further comprising steps of comparing said
inclusion criteria for a particular pool to a threshold value, and
including said particular pool in said selected set if said
threshold is met.
4. A method for managing a mortgage-backed securities index
according to claim 3, wherein said threshold value is 1.5% for all
30-year mortgage-backed securities pools.
5. A method for managing a mortgage-backed securities index
according to claim 3, wherein said threshold value is 0.4% for all
15-year mortgage-backed securities pools.
6. A method for managing a mortgage-backed securities index
according to claim 2 wherein said relative weight of said each
security within said selected set is given by the following
equation: w = x c , t .rho. a , c , t [ a = FNMA , GNMA , FHLMC c
.di-elect cons. Z t = 180 , 360 .rho. a , ct ] . ##EQU00019##
7. A method for managing a mortgage-backed securities index
according to claim 6, wherein said total return of said each
security within said selected set on any given day t.sub.2 is given
by the following equation: TR t 2 j = - p t 1 + f t 1 p t 2 + [ ( 1
- f t 1 ) + c 12 ] [ 1 + t t 1 d 360 ] - k p t 1 , ##EQU00020##
wherein p.sub.t.sub.1 is a same-day settle price of said each
security on the close of day t.sub.1, wherein p.sub.t.sub.2 is a
same-day settle price of said each security on the close of day
t.sub.2, wherein r.sub.t.sub.1 is a one-month BBA LIBOR on the
close of day t.sub.1, wherein f.sub.t.sub.1 is a monthly pay-down
factor of said each security as best determined by day t.sub.1,
said monthly pay-down factor f.sub.t.sub.1 being selected from a
sequence of monthly pay-down factors f.sub.i, t.sub.1 is the last
business day of the preceding month, t.sub.2 is any day of the
current month, and wherein d = { 25 - day of the month of t 2 , if
said security is issued by FNMA 15 - day of the month of t 2 , if
said security is issued by GNMA or FHLMC and k = { + 1 , if d >
0 - 1 , if d < 0 ##EQU00021##
8. A method for managing a mortgage-backed securities index
according to claim 7, wherein said sequence of monthly pay-down
factors is given by the following equation: f i = .alpha..alpha.
.di-elect cons. A .rho. .alpha. , i - .alpha..alpha. .di-elect
cons. A .rho. .alpha. , i + 1 .alpha..alpha. .di-elect cons. A
.rho. .alpha. , i , ##EQU00022## wherein .rho..sub..alpha.,i+1 is
the principal outstanding of pool .alpha. as of the first of month
i, and wherein .rho..sub..alpha.,i+1 is the principal outstanding
of pool .alpha. as of the first of month i+1.
9. A method for managing a mortgage-backed securities index
according to claim 8 wherein said same-day settle price is given by
the following equation: p t = p ^ t + c 12 d 1 30 1 + r d 2 360 f t
+ ( 1 - f t ) + c 12 1 + r d 3 360 , ##EQU00023## wherein {tilde
over (p)}.sub.t is the TBA price of said security 1-month forward
standard PSA settle on the close of business on day t; wherein
d.sub.1 is the number of days into the month that 1-month forward
standard PSA settle occurs; wherein d.sub.2 is the number of days
in between the purchase date and the standard PSA settle date
1-month forward inclusive of the former and exclusive of the
latter; wherein, for FNMA mortgage-backed securities, d.sub.3 is
the number of days between the purchase date and the 25th of the
next month; wherein for GNMA or FHLMC mortgage-backed securities,
d.sub.3 is the number of days between the purchase date and the
15th of the next month; and wherein r is a one-month BBA LIBOR on
the close of day t.
10. A method for managing a mortgage-backed securities index
according to claim 9 wherein said total return of said index from
day t.sub.1 in month k to day t.sub.2 in month n is given by the
following equation: TR t 1 t 2 = ( 1 + TR k - TR t 1 ) [ i = k + 1
n - 1 ( 1 + TR i ) ] ( 1 + TR t 2 ) - 1 , TR t 1 ##EQU00024## is
the month-to-date total return of said index on the day t.sub.1,
and TR.sub.t.sub.2 is the moth-to-date total return of said index
on the day t.sub.2, wherein TR.sub.k is the total return of the
index for the month k, and wherein TR.sub.i is the total return of
the index for any intermediate month between k and n.
11. A method for managing a mortgage-backed securities index
according to claim 10 wherein said total return of the index for
any intermediate month is given by the following equation: TR i = j
= 1 n w i j p ii j TR i j j = 1 n w i j p i j , ##EQU00025##
wherein said total returns of said index on the day t.sub.1 and
t.sub.2 are given by the following equation: TR t = j = 1 n w i j p
i j TR t j j = 1 n w i j p i j , ##EQU00026## wherein
{w.sub.i.sup.j}.sub.j=1.sup.n are the relative weights of said
mortgage-backed securities within said index, and wherein
p.sub.i.sup.j is the same-day settle price for said each security
within said index.
12. A method for managing a mortgage-backed securities index
according to claim 11, further comprising a step of determining a
level of said mortgage-backed securities index, said level being
given by the following equation: P t P 00 / 00 / 00 = 1 + TR 00 /
00 / 00 t , ##EQU00027## wherein P.sub.00/00/00 is the starting
level of said index, wherein TR|.sub.00/00/00.sup.t is the total
return of said index from start to the day t, and wherein P.sub.t
is the current level of the index.
13. A method for managing a mortgage-backed securities index
according to claim 12, wherein said starting level of said index is
100.
14. A method for managing a mortgage-backed securities index
according to claim 1, further comprising a step of rebalancing said
index by repeating said steps of selecting a set of mortgage-backed
securities to be included in said mortgage-backed securities index,
assigning said relative weight to each security within said
selected set, and calculating said total return of said
mortgage-backed securities index.
15. A system for managing a mortgage-backed securities index,
comprising: input device for inputting market data into said
system, said market data comprising data for all outstanding
mortgage-backed securities; a processor operating under a set of
instructions, the set of instructions including a first set of
instructions causing the processor to select a set of
mortgage-backed securities to be included in said mortgage-backed
securities index, said set of mortgaged-backed securities being
selected from said all outstanding mortgage-backed securities, the
first set of instructions further causing the processor to
aggregate said all outstanding mortgage-backed securities into a
plurality of pools, each of said pools comprising mortgage-backed
securities having the same coupon and the same original term, and
to determine inclusion criteria for each pool within said plurality
of pools, said inclusion criteria being determined using an
objective condition; a second set of instructions causing the
processor to assign a relative weight to each security within said
selected set, said relative weight being a relative proportion of
total outstanding principal on said each security to a total
outstanding principal on all securities within said selected set;
and a third set of instructions causing the processor to calculate
a total return of said mortgage-backed securities index, said total
return being calculated based on said assigned relative weight of
said each security within said selected set, and a total return of
said each security within said selected set based on a
same-day-settle price an output device outputting the calculated
total return of said mortgage-backed securities index to a
user.
16. A system for managing a mortgage-backed securities index
according to claim 15 further comprising a storage device, said
storage device storing data circulated within said system.
17. A system for managing a mortgage-backed securities index
according to claim 15, wherein the set of instructions further
comprises a forth set of instructions causing the processor to
classify said data for all outstanding mortgage-backed securities
in accordance with a coupon value, issuing agency and original term
of each of said outstanding mortgage-backed securities.
18. A system for managing a mortgage-backed securities index
according to claim 15 wherein said inclusion criterion is given by
the following equation: x c , t = [ a = { FNMA GNMA FHLMC } .rho. a
, c , t ] [ a = FNMA , GNMA , c .di-elect cons. Z t = 180 , 360
FHLMC .rho. a , c , t ] ##EQU00028## wherein .rho..sub.a,c,t is the
total outstanding principal on said outstanding mortgage-backed
securities, a is an agency which issued said outstanding
mortgage-backed securities, c is a coupon value of said outstanding
mortgage-backed securities, and t is an original term of said
outstanding mortgage-backed securities.
19. A system for managing a mortgage-backed securities index
according to claim 15, wherein the set of instructions further
comprises a fifth set of instructions causing the processor to
compare said inclusion criterion for each of said aggregated pools
to a threshold value, and to include a particular aggregated pool
in said selected set if said threshold is met.
20. A system for managing a mortgage-backed securities index
according to claim 19, wherein said threshold value is 1.5% for all
30-year mortgage-backed securities pools.
21. A system for managing a mortgage-backed securities index
according to claim 19, wherein said threshold value is 0.4% for all
15-year mortgage-backed securities pools.
22. A system for managing a mortgage-backed securities index
according to claim 18 wherein said relative weight of said each
security within said selected set is given by the following
equation: w = x c , t .rho. a , c , t [ a = FNMA , GNMA , c
.di-elect cons. Z t = 180 , 360 FHLMC .rho. a , c , t ] ,
##EQU00029## wherein w is said relative weight of said each
security within said selected set.
23. A system for managing a mortgage-backed securities index
according to claim 15, wherein said third set of instructions
further comprises a sixth set of instructions causing the processor
to calculate said total return of said each security within said
selected set.
24. A system for managing a mortgage-backed securities index
according to claim 23, wherein said sixth set of instructions
causes the processor to calculate said total return of said each
security within said selected set on any given day t.sub.2 in
accordance with the following equation: TR t 2 j = - p t 1 + f t 1
p t 2 + [ ( 1 - f t 1 ) + c 12 ] [ 1 + r t 1 d 360 ] - k p t 1 ,
##EQU00030## wherein p.sub.t.sub.1 is a same-day settle price of
said each security on the close of day t.sub.1, wherein
p.sub.t.sub.2 is a same-day settle price of said each security on
the close of day t.sub.2, wherein r.sub.t.sub.1 is a one-month BBA
LIBOR on the close of day t.sub.1, wherein f.sub.t.sub.1 is a
monthly pay-down factor of said each security as best determined by
day t.sub.1, said monthly pay-down factor f.sub.t.sub.1 being
selected from a sequence of monthly pay-down factors f.sub.i,
t.sub.1 is the last business day of the preceding month, t.sub.2 is
any day of the current month, and wherein d = { 25 - day of the
month of t 2 , if said each security is issued by FNMA 15 - day of
the month of t 2 , if said each security is issued by GNMA or FHLMC
and k = { + 1 , if d > 0 - 1 , if d < 0 ##EQU00031##
25. A system for managing a mortgage-backed securities index
according to claim 24, wherein said sequence of monthly pay-down
factors is given by the following equation: f i = .alpha..alpha.
.di-elect cons. A .rho. a , i - .alpha..alpha. .di-elect cons. A
.rho. a , i + 1 .alpha..alpha. .di-elect cons. A .rho. a , i ,
##EQU00032## wherein .rho..sub..alpha.,i is the principal
outstanding of pool .alpha. as of the first of month i, and wherein
.rho..sub..alpha.,i+1 is the principal outstanding of pool .alpha.
as of the first of month i+1.
26. A system for managing a mortgage-backed securities index
according to claim 24, wherein said same-day settle price is given
by the following equation: p t = p ~ t + c 12 d 1 30 1 + r d 2 360
f t + ( 1 - f t ) + c 12 1 + r d 3 360 , ##EQU00033## wherein
{tilde over (p)}.sub.t is the TBA price of said security 1-month
forward standard PSA settle on the close of business on day t;
wherein d.sub.1 is the number of days into the month that 1-month
forward standard PSA settle occurs; wherein d.sub.2 is the number
of days in between the purchase date and the standard PSA settle
date 1-month forward inclusive of the former and exclusive of the
latter; wherein, for FNMA mortgage-backed securities, d.sub.3 is
the number of days between the purchase date and the 25th of the
next month; wherein for GNMA or FHLMC mortgage-backed securities,
d.sub.3 is the number of days between the purchase date and the
15th of the next month; and wherein r is a one-month BBA LIBOR on
the close of day t.
27. A system for managing a mortgage-backed securities index
according to claim 24, wherein said total return of said index from
day t.sub.1 in month k to day t.sub.2 in month n is given by the
following equation: TR t 1 t 2 = ( 1 + TR k - TR t 1 ) [ i = k + 1
n - 1 ( 1 + TR i ) ] ( 1 + TR t 2 ) - 1 ##EQU00034## when k<n-1,
wherein TR.sub.t.sub.1 is the month-to-date total return of said
index on the day t.sub.1, and TR.sub.t.sub.2 is the moth-to-date
total return of said index on the day t.sub.2, wherein TR.sub.k is
the total return of the index for the month k, and wherein TR.sub.i
is the total return of the index for any intermediate month between
k and n.
28. A system for managing a mortgage-backed securities index
according to claim 27, wherein said total return of the index for
any intermediate month is given by the following equation: TR i = j
= 1 n w i j p ii j TR i j j = 1 n w i j p i j , ##EQU00035##
wherein said total returns of said index on the day t.sub.1 and
t.sub.2 are given by the following equation: TR t = j = 1 n w i j p
i j TR t j j = 1 n w i j p i j , ##EQU00036## wherein
{w.sub.i.sup.j}.sub.j=1.sup.n are the relative weights of said
mortgage-backed securities within said index, and wherein
p.sub.i.sup.j is the same-day settle price for said each security
within said index.
29. A system for managing a mortgage-backed securities index
according to claim 24, wherein said total return of said index from
day t.sub.1 in month k to day t.sub.2 in month n is given by the
following equation:
TR|.sub.t.sub.1.sup.t.sup.2=TR.sub.t.sub.2-TR.sub.t.sub.1 when k=n,
wherein TR.sub.t.sub.1 is the total return of said index on the day
t.sub.1, and TR.sub.t.sub.2 is the total return of said index on
the day t.sub.2.
30. A system for managing a mortgage-backed securities index
according to claim 24 wherein said total return of said index from
day t.sub.1 in month k to day t.sub.2 in month n is given by the
following equation:
TR|.sub.t.sub.1.sup.t.sup.2=(1+TR.sub.k-TR.sub.t.sub.1)(1+TR.sub.t.sub.2)-
-1 when k=n-1, wherein TR.sub.t.sub.1 is the total return of said
index on the day t.sub.1, wherein TR.sub.t.sub.2 is the total
return of said index on the day t.sub.2, and wherein TR.sub.k is
the total return of the index for the month k.
31. A system for managing a mortgage-backed securities index
according to claim 15, wherein the set of instructions further
comprises a seventh set of instructions configured to determine a
level of said mortgage-backed securities index.
32. A system for managing a mortgage-backed securities index
according to claim 31 wherein said level is given by the following
equation: P t P 00 / 00 / 00 = 1 + TR 00 / 00 / 00 t , ##EQU00037##
wherein P.sub.t is said level of said mortgage-backed securities
index on the day t, wherein P.sub.00/00/00 is a starting level of
said index, and wherein TR|.sub.00/00/00.sup.t is the total return
of said index from the starting date to the day t.
33. A system for managing a mortgage-backed securities index
according to claim 32, wherein said starting level of said index is
100.
34. A mortgage-backed securities index, comprising: a set of
mortgage-backed securities, said set of mortgaged-backed securities
being selected from all outstanding mortgage-backed securities,
wherein a relative weight is assigned to each security within said
selected set, said relative weight being a relative proportion of
total outstanding principal on said each security to the total
outstanding principal on all securities within said selected set,
wherein said mortgage-backed securities index is characterized by a
total return of said mortgage-backed securities index, said total
return being calculated based on said assigned relative weight for
said each security, and a total return of said each security based
on a same-day-settle price, wherein said selected set of
mortgage-backed securities is selected by aggregating said all
outstanding mortgage-backed securities into a plurality of pools,
each of said pools comprising mortgage-backed securities having the
same coupon and the same original term, and calculating an
inclusion criteria for each pool within said plurality of pools,
and wherein said inclusion criteria is determined using an
objective condition.
35. A mortgage-backed securities index according to claim 34,
wherein said inclusion criteria is given by the following equation:
x c , t = [ a = { FNMA GNMA FHLMC } .rho. a , c , t ] [ a = FNMA ,
GNMA , c .di-elect cons. Z t = 180 , 360 FHLMC .rho. a , c , t ]
##EQU00038## wherein .rho..sub.a,c,t is the total outstanding
principal on said outstanding mortgage-backed securities, a is an
agency which issued said outstanding mortgage-backed securities, c
is a coupon value of said outstanding mortgage-backed securities,
and t is an original term of said outstanding mortgage-backed
securities.
36. A mortgage-backed securities index according to claim 35,
wherein if said inclusion criteria for a particular pool is greater
than a threshold value, said particular pool is included in said
selected set.
37. A mortgage-backed securities index according to claim 36,
wherein said threshold value is 1.5% for all 30-year
mortgage-backed securities pools.
38. A mortgage-backed securities index according to claim 36,
wherein said threshold value is 0.4% for all 15-year
mortgage-backed securities pools.
39. A mortgage-backed securities index according to claim 35
wherein said relative weight of said each security within said
selected set is given by the following equation: w = x c , t .rho.
a , c , t [ a = FNMA , GNMA , c .di-elect cons. Z t = 180 , 360
FHLMC .rho. a , c , t ] , ##EQU00039##
40. A mortgage-backed securities index according to claim 39,
wherein said total return of said each security within said
selected set on any given day t.sub.2 is given by the following
equation: TR t 2 j = - p t 1 + f t 1 p t 2 + [ ( 1 - f t 1 ) + c 12
] [ 1 + r t 1 d 360 ] - k p t 1 , ##EQU00040## wherein
p.sub.t.sub.1 is a same-day settle price of said each security on
the close of day t.sub.1, wherein p.sub.t.sub.2 is a same-day
settle price of said each security on the close of day t.sub.2,
wherein r.sub.t.sub.1 is a one-month BBA LIBOR on the close of day
t.sub.1, wherein f.sub.t.sub.1 is a monthly pay-down factor of said
each security as best determined by day t.sub.1, said monthly
pay-down factor f.sub.t.sub.1 being selected from a sequence of
monthly pay-down factors f.sub.i, t.sub.1 is the last business day
of the preceding month, t.sub.2 is any day of the current month,
and wherein d = { 25 - day of the month of t 2 , if said security
is issued by FNMA 15 - day of the month of t 2 , if said security
is issued by GNMA or FHLMC and k = { + 1 , if d > 0 - 1 , if d
< 0 ##EQU00041##
41. A mortgage-backed securities index according to claim 40,
wherein said sequence of monthly pay-down factors is given by the
following equation: f i = .alpha..alpha. .di-elect cons. A .rho. a
, i - .alpha..alpha. .di-elect cons. A .rho. a , i + 1
.alpha..alpha. .di-elect cons. A .rho. a , i , ##EQU00042## wherein
.rho..sub..alpha.,i is the principal outstanding of pool .alpha. as
of the first of month i, and wherein .rho..sub..alpha.,i+1 is the
principal outstanding of pool .alpha. as of the first of month
i+1.
42. A mortgage-backed securities index according to claim 41,
wherein said same-day settle price is given by the following
equation: p t = p ~ t + c 12 d 1 30 1 + r d 2 360 f t + ( 1 - f t )
+ c 12 1 + r d 3 360 , ##EQU00043## wherein {tilde over (p)}.sub.t
is the TBA price of said security 1-month forward standard PSA
settle on the close of business on day t; wherein d.sub.1 is the
number of days into the month that 1-month forward standard PSA
settle occurs; wherein d.sub.2 is the number of days in between the
purchase date and the standard PSA settle date 1-month forward
inclusive of the former and exclusive of the latter; wherein, for
FNMA mortgage-backed securities, d.sub.3 is the number of days
between the purchase date and the 25th of the next month; wherein
for GNMA or FHLMC mortgage-backed securities, d.sub.3 is the number
of days between the purchase date and the 15th of the next month;
and wherein r is a one-month BBA LIBOR on the close of day t.
43. A mortgage-backed securities index according to claim 42,
wherein said total return of said index from day t.sub.1 in month k
to day t.sub.2 in month n is given by the following equation: TR |
t 1 t 2 = ( 1 + TR k - TR t 1 ) [ i = k + 1 n - 1 ( 1 + TR i ) ] (
1 + TR t 2 ) - 1 , ##EQU00044## wherein TR.sub.t.sub.1 is the
month-to-date total return of said index on the day t.sub.1, and
TR.sub.t.sub.2 is the moth-to-date total return of said index on
the day t.sub.2, wherein TR.sub.k is the total return of the index
for the month k, and wherein TR.sub.i is the total return of the
Index for any intermediate month between k and n.
44. A mortgage-backed securities index according to claim 43
wherein said total return of the index for any intermediate month
is given by the following equation: TR i = j = 1 n w i j p ii j TR
i j j = 1 n w i j p i j , ##EQU00045## wherein said total returns
of said index on the day t.sub.1 and t.sub.2 are given by the
following equation: TR t = j = 1 n w i j p i j TR t j j = 1 n w i j
p i j , ##EQU00046## wherein {w.sub.i.sup.j}.sub.j=1.sup.n are the
relative weights of said mortgage-backed securities within said
index, and wherein p.sub.i.sup.j is the same-day settle price for
said each security within said index.
45. A mortgage-backed securities index according to claim 44,
wherein said index further comprises a level, said level being
given by the following equation: P t P 00 / 00 / 00 = 1 + TR | 00 /
00 / 00 t , ##EQU00047## wherein P.sub.00/00/00 is the starting
level of said index, wherein TR|.sub.00/00/00.sup.t is the total
return of said index from start to the day t, and wherein P.sub.t
is the current level of the index.
46. A mortgage-backed securities index according to claim 45,
wherein said starting level of said index is 100.
47. A mortgage-backed securities index according to claim 34,
wherein said index is rebalanced by of selecting a new set of
mortgage-backed securities to be included in said mortgage-backed
securities index, assigning said relative weight to each security
within said new selected set, and calculating a new total return of
said mortgage-backed securities index.
48. A mortgage-backed securities index according to claim 47
wherein said index is rebalanced on a last business day of each
month.
49. A computer program for managing a mortgage-backed securities
index executable on general purpose computer, comprising: an input
segment for inputting market data into said system, said market
data comprising data for all outstanding mortgage-backed
securities; a selection segment for selecting a set of
mortgage-backed securities to be included in said mortgage-backed
securities index, said set of mortgaged-backed securities being
selected from said all outstanding mortgage-backed securities; a
weight segment for assigning a relative weight to each security
within said selected set, said relative weight being a relative
proportion of total outstanding principal on said each security to
a total outstanding principal on all securities within said
selected set; a total return segment for calculating a total return
of said mortgage-backed securities index, said total return being
calculated based on said assigned relative weight of said each
security within said selected set, and a total return of said each
security within said selected set based on a same-day-settle price;
a classification segment, said classification segment classifying
said data for all outstanding mortgage-backed securities in
accordance with a coupon value, issuing agency and original term of
each of said outstanding mortgage-backed securities; and an
aggregation segment, said aggregation segment aggregating said
outstanding mortgage-backed securities into a plurality of
aggregated pools, wherein said aggregation segment further
comprises a segment for calculating an inclusion criterion for each
of said aggregated pools, said inclusion criterion being determined
using an objective condition.
50. A computer program for managing a mortgage-backed securities
index according to claim 49 further comprising a storage segment,
said storage segment storing data circulated within said
system.
51. A computer program for managing a mortgage-backed securities
index according to claim 49 wherein said inclusion criterion is
given by the following equation: x c , t = [ a = { FNMA GNMA FHLMC
} .rho. a , c , t ] [ a = FNMA , GNMA , FHLMC c .di-elect cons. Z t
= 180 , 360 .rho. a , c , t ] ##EQU00048## wherein .rho..sub.a,c,t
is the total outstanding principal on said outstanding
mortgage-backed securities, a is an agency which issued said
outstanding mortgage-backed securities, c is a coupon value of said
outstanding mortgage-backed securities, and t is an original term
of said outstanding mortgage-backed securities.
52. A computer program for managing a mortgage-backed securities
index according to claim 51, further comprising a segment for
comparing said inclusion criterion for all of said aggregated pools
to a threshold value, and including an aggregated pool in said
selected set if said threshold is met.
53. A computer program for managing a mortgage-backed securities
index according to claim 52, wherein said threshold value is 1.5%
for all 30-year mortgage-backed securities pools.
54. A computer program for managing a mortgage-backed securities
index according to claim 52, wherein said threshold value is 0.4%
for all 15-year mortgage-backed securities pools.
55. A computer program for managing a mortgage-backed securities
index according to claim 51 wherein said relative weight of said
each security within said selected set is given by the following
equation: w = x c , t .rho. a , c , t [ a = FNMA , GNMA , FHLMC c
.di-elect cons. Z t = 180 , 360 .rho. a , c , t ] , ##EQU00049##
wherein w is said relative weight of said each security within said
selected set.
56. A computer program for managing a mortgage-backed securities
index according to claim 51 wherein said total return segment
further comprises segment for calculating said total return of said
each security within said selected set.
57. A computer program for managing a mortgage-backed securities
index according to claim 56 wherein said segment for calculating
calculates said total return of said each security within said
selected set on any given day t.sub.2 in accordance with the
following equation: TR t 2 j = - p t 1 + f t 1 p t 2 + [ ( 1 - f t
1 ) + c 12 ] [ 1 + r t 1 d 360 ] - k p t 1 , ##EQU00050## wherein
p.sub.t.sub.1 is a same-day settle price of said each security on
the close of day t.sub.1, wherein p.sub.t.sub.2 is a same-day
settle price of said each security on the close of day t.sub.2,
wherein r.sub.t.sub.1 is a one-month BBA LIBOR on the close of day
t.sub.1, wherein f.sub.t.sub.1 is a monthly pay-down factor of said
each security as best determined by day t.sub.1, said monthly
pay-down factor f.sub.t.sub.1 being selected from a sequence of
monthly pay-down factors f.sub.i, t.sub.1 is the last business day
of the preceding month, t.sub.2 is any day of the current month,
and wherein d = { 25 - day of the month of t 2 , if said each
security is issued by FNMA 15 - day of the month of t 2 , if said
each security is issued by GNMA or FHLMC and k = { + 1 , if d >
0 - 1 , if d < 0 ##EQU00051##
58. A computer program for managing a mortgage-backed securities
index according to claim 57, wherein said sequence of monthly
pay-down factors is given by the following equation: f i = .alpha.
o .di-elect cons. A .rho. .alpha. , i - .alpha. o .di-elect cons. A
.rho. .alpha. , i + 1 .alpha. o .di-elect cons. A .rho. .alpha. , i
, ##EQU00052## wherein .rho..sub..alpha.,i is the principal
outstanding of pool .alpha. as of the first of month i, and wherein
.rho..sub..alpha.,i+1 is the principal outstanding of pool .alpha.
as of the first of month i+1.
59. A computer program for managing a mortgage-backed securities
index according to claim 58, wherein said same-day settle price is
given by the following equation: p t = p ~ t + c 12 d 1 30 1 + r d
2 360 f t + ( 1 - f t ) + c 12 1 + r d 3 360 , ##EQU00053## wherein
{tilde over (p)}.sub.t is the TBA price of said security 1-month
forward standard PSA settle on the close of business on day t;
wherein d.sub.1 is the number of days into the month that 1-month
forward standard PSA settle occurs; wherein d.sub.2 is the number
of days in between the purchase date and the standard PSA settle
date 1-month forward inclusive of the former and exclusive of the
latter; wherein, for FNMA mortgage-backed securities, d.sub.3 is
the number of days between the purchase date and the 25th of the
next month; wherein for GNMA or FHLMC mortgage-backed securities,
d.sub.3 is the number of days between the purchase date and the
15th of the next month; and wherein r is a one-month BBA LIBOR on
the close of day t.
60. A computer program for managing a mortgage-backed securities
index according to claim 51 wherein said total return of said index
from day t.sub.1 in month k to day t.sub.2 in month n is given by
the following equation: TR | t 1 t 2 = ( 1 + TR k - TR t 1 ) [ i =
k + 1 n - 1 ( 1 + TR i ) ] ( 1 + TR t 2 ) - 1 ##EQU00054## when
k<n-1, wherein TR.sub.t.sub.1 is the month-to-date total return
of said index on the day t.sub.1, and TR.sub.t.sub.2 is the
moth-to-date total return of said index on the day t.sub.2, wherein
TR.sub.k is the total return of the Index for the month k, and
wherein TR.sub.i is the total return of the index for any
intermediate month between k and n.
61. A computer program for managing a mortgage-backed securities
index according to claim 60, wherein said total return of the index
for any intermediate month is given by the following equation: TR i
= j = 1 n w i j p ii j TR i j j = 1 n w i j p i j , ##EQU00055##
wherein said total returns of said index on the day t.sub.1 and
t.sub.2 are given by the following equation: TR t = j = 1 n w i j p
i j TR t j j = 1 n w i j p i j , ##EQU00056## wherein
{w.sub.i.sup.j}.sub.j=1.sup.n are the relative weights of said
mortgage-backed securities within said index, and wherein
p.sub.i.sup.j is the same-day settle price for said each security
within said index.
62. A computer program for managing a mortgage-backed securities
index according to claim 51 wherein said total return of said index
from day t.sub.1 in month k to day t.sub.2 in month n is given by
the following equation:
TR|TR.sub.t.sub.1.sup.t.sup.2=TR.sub.t.sub.2-TR.sub.t.sub.1 when
k=n, wherein TR.sub.t.sub.1 is the total return of said index on
the day t.sub.1, and TR.sub.t.sub.2 is the total return of said
index on the day t.sub.2.
63. A computer program for managing a mortgage-backed securities
index according to claim 51 wherein said total return of said index
from day t.sub.1 in month k to day t.sub.2 in month n is given by
the following equation:
TR|.sub.t.sub.1.sup.t.sup.2=(1+TR.sub.k-TR.sub.t.sub.1)(1+TR.sub.t.sub.2)-
-1 when k=n-1, wherein TR.sub.t.sub.1 is the total return of said
index on the day t.sub.1, wherein TR.sub.t.sub.2 is the total
return of said index on the day t.sub.2 and wherein TR.sub.k is the
total return of the index for the month k.
64. A computer program for managing a mortgage-backed securities
index according to claim 49, further comprising level segment, said
level segment determining a level of said mortgage-backed
securities index.
65. A computer program for managing a mortgage-backed securities
index according to claim 64 wherein said level is given by the
following equation: P t P 00 / 00 / 00 = 1 + TR | 00 / 00 / 00 t ,
##EQU00057## wherein P.sub.t is said level of said mortgage-backed
securities index on the day t, wherein P.sub.00/00/00 is a starting
level of said index, and wherein TR|.sub.00/00/00.sup.t is the
total return of said index from the starting date to the day t.
66. A computer program for managing a mortgage-backed securities
index according to claim 65, wherein said starting level of said
index is 100.
67. A computer program for managing a mortgage-backed securities
index according to claim 49 further comprising an output segment,
said output segment displaying said level of said mortgage-backed
securities index and said total return of said mortgage-backed
securities index to the user.
Description
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] This application is a continuation of U.S. patent
application Ser. No. 09/931,149 filed Aug. 16, 2001, entitled
"METHOD AND SYSTEM FOR MANAGING A MORTGAGE-BACKED SECURITIES
INDEX", in the name of David Rios et al. the entire disclosure of
which application is hereby incorporated by reference.
BACKGROUND OF THE INVENTION
[0002] 1. Field of the Invention
[0003] The invention generally relates to a method and system for
generating an Index-type benchmark for a mortgage-backed securities
(MBS) sector of the securities market.
[0004] 2. Description of the Related Art
[0005] In the stock market, an index is a device that measures
changes in the prices of a basket of shares, and represents the
changes using a single Fig. The purpose is to give investors an
easy way to see the general direction of shares in the index. The
FTSE 100 Index, for example, is calculated by taking a weighted
average of the share prices of the largest 100 companies on the
London Stock Exchange. Launched in 1984 with a base Fig. of 1,000,
the FTSE is calculated continuously throughout the trading day.
[0006] Generally speaking, an index is defined by a pre-determined
universe of individual issues valued or weighted in proportion to
their size within the whole universe of issues. The Standard &
Poor's 500 Index is an example of such an index and is based on the
common stocks of the 500 largest corporations which trade on United
States exchanges. Illustratively, the weighting for each different
security in the S & P 500 index is simply the market
capitalization of that security expressed as a percentage of the
total market capitalization taken across all the securities then in
the index. Of course, these weights continually change as share
prices move which, in turn, changes market capitalization both of
each individual security in the index as well as that across the
entire index. Other indices, such as the Value Line index, assign
their own, e.g. equal, weightings to each security therein.
[0007] Capitalization weighting is the most common approach to
equity indices and equity portfolios. Cap-weighted portfolios are
easy to maintain because the rebalancing process takes care of
itself. An algorithm for rebalancing a capitalization weighted
stock index is disclosed, for example, in U.S. Pat. No. 6,061,663
(the '663 patent). The '663 patent teaches a method and system for
rebalancing an equity portfolio based on weighting characteristics
of individual securities. Similar to other known equity indices,
e.g. S&P 500 Index, in the '663 patent the weighting
characteristic of a particular security is based on the market
capitalization of that security expressed as a percentage of the
total market capitalization taken across all the securities then in
the index. The computer program disclosed in the '663 patent
includes instructions to classify stocks in the index into several
categories by comparing their capitalization weight to a threshold
level.
[0008] Similarly, U.S. Pat. No. 5,819,238 (the '238 patent)
discloses a method for automatically modifying a financial
portfolio having a pre-defined universe of securities, such as an
index fund that tracks a given capitalization weighted index,
through dynamic re-weighting of a position held in each such
security. Specifically, in the disclosed computer system, a target
weight is associated with each such security relative to others in
the same portfolio in proportion to a non-constant function of
current capitalization weights of the securities in the index. Once
these target weights are determined, then, in response to both the
target weight of each such security and an actual weight, as a
proportion of the portfolio in which that security was held, a
trade will be generated by the system in order to conform, within a
predefined band, the actual weight to the target weight so as to
rebalance the holdings in the portfolio.
[0009] For indices reflecting international securities markets,
relative weights may be calculated based on countries' relative
GDPs or imports.
[0010] Indices are currently used in the industry as benchmarks
allowing investors and portfolio managers to compare performance of
one sector of the securities market to others. Indices are often
used in order to create index funds, i.e., funds that purchase
securities that mimic or represent a specific index, for example
the Vanguard 500 Index Fund mimics the composition and, supposedly,
performance of the S & P 500 stock index.
[0011] Mortgage-Backed Securities are securities backed by mortgage
loans, including pass-through securities, modified pass-through
securities, mortgage-backed bonds, and mortgage pay-through
securities. MBS are created when mortgage loans are pooled and
underwritten by eligible issuers. Commonly referred to as
"pass-through" certificates, these MBS entitle an investor to an
undivided interest in the underlying mortgage loan pool. Thus, an
investor receives a pro rata share of the interest (net of
servicing and guaranty fees) and/or principal on the underlying
mortgage loans.
[0012] Several financial institutions have developed indexes for
measuring changes in the MBS markets. For example, Lehman Brothers
has developed an MBS Index (hereinafter "LB MBS Index") which
covers the mortgage-backed pass-through securities of GNMA (also
known as "Ginnie Mae"), FNMA (also known as "Fannie Mae"), and
FHLMC (also known as "Freddie Mac"). It is formed by grouping the
universe of over 600,000 individual fixed rate MBS pools into
approximately 3,500 generic aggregates. The aggregates included are
priced daily using a matrix pricing routine based on trade price
quotations by agency, program, coupon, and degree of seasoning.
Lehman Brothers also developed a Mortgage-Backed Securities Index
which is an unmanaged version of the LB MBS index and is composed
of all fixed securities mortgage pools by GNMA, FNMA and the FHLMC,
including GNMA Graduated Payment Mortgages.
[0013] In order to generate one of the LB MBS indices, a user must
first select a set of securities that satisfy a number of
subjective rules (for example, the total outstanding balance of
each generic security must be at least $100 million), then price
each issue within the index based on the provided "matrix pricing,"
calculate the returns of each individual security within the index
and calculate the index return as a market-weighted average of
individual security returns. The weights of individual securities
are subjectively assigned and are not related to the proportion of
the total outstanding principal on a particular security and total
outstanding principal for the selected pool. Additionally, as
explained further below, mortgage-backed securities are traded in
"to-be-announced" (TBA) transactions where the purchase price is
settled at some future TBA date. In order to calculate an index's
total return, TBA settle prices for each security have to be
converted into same-day-settle prices. According to the LB MBS
Index's published algorithm, this process involves complicated
calculations which include adjustments for an unknown future cash
flow.
SUMMARY OF THE INVENTION
[0014] The present invention is directed to a method and system for
generating an MBS Index by objectively selecting securities from
all available mortgage-backed securities, assigning a relative
weight to each selected security, evaluating the MBS Index by
calculating its total return and periodically rebalancing the
Index. There is a need in the industry for a system and method for
generating an MBS Index, which would objectively select
mortgage-based securities from the entire universe of available
securities based on an easy-to-administer mathematical algorithm.
There is also a need for a simple and objective system and method
for calculating total return of the MBS Index based on
same-day-settle price of each included security that would
eliminate any guess work as to the future cash flow.
[0015] It is an object of the present invention to provide a system
and method for generating an objective benchmark, which will
accurately reflect the value of the MBS sector of the securities
market.
[0016] It is another object of the present invention to provide a
system and method for generating an MBS Index, which will take into
consideration all outstanding mortgage-backed securities.
[0017] It is a further object of the present invention to provide a
system and method for managing an MBS index, which can be easily
automatically rebalanced.
[0018] It is still another object of the present invention to
provide a system and method for generating and managing an MBS
Index, which will allow portfolio managers to more precisely
measure the performance of the mortgage-backed securities sector
relative to other fixed-income investments.
[0019] It is still a further object of the present invention to
provide a system and a method that allow portfolio managers to
create mutual or exchange-traded funds, which will purchase, hold
and sell mortgage-backed securities mimicking and/or representing
the provided MBS Index.
[0020] In accordance with the preferred embodiment of the present
invention, a system and method for generating and managing an MBS
Index are provided. In order to determine which securities will be
represented in the Index during a particular month, the system and
method of the present invention perform a set of calculations
during the second business week of the preceding month. For the
purpose of selecting securities for the Index, all outstanding
mortgage-backed securities are considered. They are preferably
aggregated into pools based on their coupon and original term (e.g.
15 and 30 years). If the total principal outstanding for any
30-year coupon represents more than 1.5% of the total, then this
30-year coupon will be included in the Index. Similarly, if the
total principal outstanding on any 15-year coupon represents more
than 0.4% of the total, then this 15-year coupon will be included
in the Index.
[0021] The performance of the Index is measured by its total
return. The system and method of the invention calculate the total
return of the generic Index in accordance with an algorithm which
is provided herein. In the preferred embodiment, the total return
of the Index depends on a total return of each security included in
the index weighted according to the relative weight assigned to
such particular security. The present invention provides a method
of assigning relative weights, which represent relative proportions
of different generic securities in the generic MBS Index, and
covers the frequency of reweighing. In accordance with another
embodiment of the present invention, the system and method may be
modified to produce the relative weights of the individual
securities within Conventional, Government, 30-year, and 15-year
indices.
[0022] The above and other objects, aspects, features and
advantages of the invention will be more readily apparent from the
description of the preferred embodiments thereof taken in
conjunction with the accompanying drawings and appended claims.
BRIEF DESCRIPTION OF THE DRAWINGS
[0023] The invention is illustrated by way of the example described
below and the FIGS. of the accompanying drawings in which like
references denote like or corresponding parts, which in no way
should be considered as a limitation of the invention.
[0024] FIG. 1 is a block diagram of the system 50 of the present
invention;
[0025] FIG. 2 is a functional diagram of the discrete components
and information flow within the system for generating and managing
an MBS Index in accordance with the present invention;
[0026] FIG. 3 is the table of The Bond Market Association MBS
Notification and Settlement Dates from May 2001 to August 2001, the
list was last updated on Apr. 27, 2001;
[0027] FIG. 4 is a schematic flow chart of the steps to be
performed in the method of the present invention;
[0028] FIG. 5 is a logic flow chart of the processing logic for the
index composition determination;
[0029] FIG. 6 is a logic flow chart of the processing logic for the
relative weights calculation;
[0030] FIG. 7 is a logic flow chart of the processing logic for the
pay-down factor determination;
[0031] FIG. 8 is a logic flow chart of the processing logic for the
same-day-settle price conversion; and
[0032] FIG. 9 is a logic flow chart of the processing logic for the
determination of total return of the MBS Index.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS AND THE
DRAWINGS
[0033] In accordance with the preferred embodiment of the present
invention, a system and method for generating, evaluating and
managing different Mortgage-Backed Securities Indices (hereinafter
MBS Indices) are provided. The method of the invention is
accomplished by a system, as shown in FIGS. 1 and 2 (described
further below), which may be implemented by a general purpose
computer operating in accordance with the instructions provided by
the system. The system 50 is illustrated in the diagram of FIG. 1
and includes a central processing unit (CPU) 1, memory comprising
ROM 4, for permanently storing program instructions, operating
parameters, or other data required for system operation, and RAM 5,
for the temporary storage and manipulation of data during the
operation of the system 50. The system further includes an input
device 3 which may be a keyboard, a handwriting recognition device,
a voice recognition device or any other known device for inputting
data including any type of hardware or software suitable for
inputting data from another computer system or the Internet. The
above described input devices are all designated in the diagram of
FIG. 1 with a reference number 3. The input device 3 inputs data,
operational parameters and commands to the CPU 1 through an
input/output interface 2. Data is output from the CPU 1 through
another input/output interface 6 to the output device 7 which may
be a printer, a computer display, a modem, or any other known
output device. As described in more detail below, the system may
comprise a plurality of discrete components which perform the
functions of selecting mortgage-backed securities to be included in
the Index, assigning relative weight to each selected security,
calculating the total return of the Index, calculating the level of
the Index and other functions in accordance with the present
invention. In the preferred embodiment, a single CPU performs these
functions. Alternatively, separate CPUs may be used or the CPU may
be segregated to perform the described functions.
[0034] The invention is described herein in connection with MBS
indices implemented by Credit Suisse First Boston ("CSFB"), a
global investment bank providing securities underwriting, sales and
trading, investment banking, private equity, financial advisory and
other related services. There are sixty four (64) different MBS
indices currently developed by CSFB. This application discloses the
invention with regard to the generic MBS Index (i.e., TBA Mortgage
Index), unless otherwise specified. As will be apparent to one of
ordinary skill in the art, the algorithm may be modified to
generate and manage different MBS indices. This disclosure provides
examples of such possible modifications where appropriate. For
example, individual weight-calculating formulas are provided for
Conventional, Government, 30-year and 15-year indices.
[0035] FIG. 4 is a flow chart of the method of the present
invention. The method may be implemented by a general purpose
computer operating in accordance with an algorithm 10 for
generating, evaluating and rebalancing a generic MBS Index (i.e.,
TBA Mortgage Index). In accordance with the preferred embodiment of
the present invention, all outstanding mortgage-backed securities
are considered in block 12, in order to generate a generic MBS
Index. These securities are either included or excluded from the
Index based on their total principal outstanding. Equations 1.1 and
1.2 describe the conditions for inclusion and elimination of MBS
securities. The relative weights are then assigned to each included
security, in block 14, in accordance with Equation 1.3. To evaluate
the performance of the provided Index over a desired time interval,
a total return for each pass-through present in the Index during
this time interval is first determined, block 16. This total return
of each generic pass-through is calculated in accordance with
Equation 4.1. Using these total returns for all generic securities
calculated in block 16, the total return of the Index can be
determined next, at block 18, using one of the Equations 2.1, 2.2
or 2.3, depending on the length of the desired time interval.
[0036] In accordance with the preferred embodiment, each provided
Index may also be characterized by its level. The level of the
Index is based on the total return of the Index over the life of
the Index (or any other desired time interval) and is calculated in
block 20 using Equations 6.2 and 6.3. Each Index is then rebalanced
in block 22, preferably on the last business day of each month, by
repeating the above steps 12 through 20 (blocks 12-20 of FIG.
4).
[0037] Referring specifically to the operations to be performed in
blocks 12 and 14, Equations 1.1, 1.2 and 1.3 present the algorithm
for including and excluding bonds in the TBA Mortgage Index as well
as the method of assigning relative weights to the generic
pass-throughs included in the Index. As described further below,
this algorithm may be modified in the process of creating
non-generic indices, for example, the Conventional, Government,
30-year, and 15-year indices. As will be obvious to one of ordinary
skill in the art, other MBS indices may also be created based on
the provided generic algorithm.
[0038] Generic mortgage-backed securities are preferably
differentiated by agency, coupon, and original term only. Before
the relative weights are computed, the composition of the Index has
to be ascertained, i.e., it has to be determined which securities
to include in the Index and which to exclude from the Index. For
this purpose, all TBA-eligible pools (i.e., pools of
mortgage-backed securities eligible for "to-be-announced"
transactions) should be considered. TBA-eligible pools are defined
by the Bond Market Association (also referred to in the present
disclosure as PSA, i.e., Public Securities Association), and these
definitions are incorporated herein by reference. These
TBA-eligible pools will preferably include mortgage-backed
securities issued by FNMA, GNMA I (i.e., modified pass-through
mortgage-backed securities on which registered holders receive
separate principal and interest payments on each of their
certificates), GNMA II (i.e., modified pass-through mortgage-backed
securities on which registered holders receive an aggregate
principal and interest payment from a central paying agent on all
of their Ginnie Mae II MBS), and/or FHLMC, all with fixed-rate
coupons and original terms of 30 or 15 years. These pools are
preferably aggregated based solely on coupon and original term. The
original year term and the coupon value may be preprogrammed in the
system or may be input into the system as operating parameters.
Additional information related to these MBS may also be
preprogrammed or input into the system.
[0039] In accordance with the preferred embodiment, if the total
principal outstanding of any 30-year coupon represents more than
1.5% of the total outstanding principal for all considered pools,
then that 30-year coupon will be included in the Index. Likewise,
if the total principal outstanding on any 15-year coupon represents
more than 0.4% of the total outstanding principal for all
considered pools, then that 15-year coupon will be included in the
Index. For example, if 30-year coupons having value of 8.5%
represent more than 1.5% of the total principal outstanding, then
FNMA 8.5, GNMA 8.5 and FHLMC 8.5 are all included in the generic
TBA Mortgage Index. If, collectively, they represent less than 1.5%
of the total outstanding, then none of these generic securities is
included in the Index. The percentages (i.e., the 1.5% and 0.4%
thresholds values) for each coupon term may be preprogrammed in the
system or input into the system as operating parameters. In an
alternative embodiment, after the above conditions are satisfied,
the Index may include only the top six 30-year coupons and the top
six 15-year coupons, as ranked by their outstanding principals. The
number of coupons to be included in the index may also be
preprogrammed or input into the system.
[0040] The data regarding outstanding principal amounts may be
continuously streamed into the system or input into the system when
it becomes available. The above calculations are preferably
performed during the second full business week of each month, when
the agencies' data becomes available, in order to determine which
bonds to include in the Index as of the close of the last business
day of that month. It should be noted that inclusion in one month
does not guarantee inclusion in the next month. The time for
performing the above calculations (i.e., the second full business
week) may be preprogrammed in the system or input into the system
as an operating parameter.
[0041] The total principal outstanding in a TBA eligible pool
issued by agency a with a coupon c and original term t as of the
first of each month is referred to herein as .rho..sub.a,c,t, and,
as noted above, is preferably calculated in the second week of that
month for each fixed-rate agency pass through. Therefore, for each
coupon and original term, the inclusion criterion x.sub.c,t may be
defined by the following equations:
x c , 360 = 1 if x c , 360 > 1.5 % x c , 180 = 1 if x c , 180
> 0.4 % x c , t = 0 otherwise , with ( Eq . 1.1 ) x c , t = [ a
= { FNMA GNMA FHLMC } .rho. a , c , t ] [ a = FNMA , GNMA c
.di-elect cons. Z t = 180 , 360 , FHLMC .rho. a , c , t ] ( Eq .
2.1 ) ##EQU00001##
[0042] The system determines the composition of the index based on
the preprogrammed values, operating parameters and the agencies'
input principal data in accordance with the above equation to
perform step 12 of FIG. 4.
[0043] After deciding which generic mortgage-backed securities are
to be included in the Index, the relative weight of each generic
pass-through may be assigned (step 14 of FIG. 4). In accordance
with the present invention, the relative weights are based on the
relation of the current outstanding principal of the considered
generic pass-through to the total outstanding principal represented
in the Index and are preferably expressed in percentage terms.
These weights represent the relative proportions of different
generic fixed-rate agency pass-throughs in the generic Index. In
accordance with this embodiment, Equation 1.3 mathematically
defines the weight of each generic pass-through in the generic TBA
Mortgage Index as a function of agency, coupon, and original term
as:
w = x c , t .rho. a , c , t [ a = FNMA , GNMA c .di-elect cons. 7 t
= 180 , 360 , FHLMC .rho. a , c , t ] ( Eq . 1.3 ) ##EQU00002##
[0044] Although the relative weights are preferably calculated
during the second week of each month, they preferably become
effective as of the close of the last business day of the month in
which the calculation took place. The time for calculating the
relative weights and the time they become effective may be
preprogrammed in the system or input into the system as operating
parameters. To ensure reliability of the data used in the disclosed
calculations, the sources of all data pertaining to outstanding
principals should be the agencies themselves. On the last business
day of each month, a set of weights for each index is output or
otherwise published.
[0045] The above algorithm for including and excluding generic
pass-throughs and for calculating their relative weights within the
generic MBS Index may be modified to produce the weights of the
Conventional, Government, 30-year, and 15-year indices. An
operation parameter may be input into the system to signal this
modification in the provided method. For the Conventional Pass
Through Index, when computing the weights using Equation 1.3, all
weights assigned to Ginnie Mae pass-throughs are set to 0 during
the final step. Similarly, when computing the weights for the
Ginnie Mae (i.e., "Government") Pass-Through Index using Equation
1.3, all weights assigned to non-Ginnie Mae pass-throughs are set
to 0 during the final step. When computing the weights for the
30-Year Pass-Through Index using the above algorithm, all weights
assigned to 180-month original term pass-throughs are set to 0
during the final step. Similarly, when computing the weights for
the 15-Year Pass-Through Index using the above algorithm, all
weights assigned to 360-month original term pass-throughs are set
to 0 during the final step. Therefore, when total returns of these
indices are calculated, total returns from the excluded securities
will be equal to "0," and only returns from the included securities
will be considered and outputted from the system. Thus, even though
all outstanding mortgage-backed securities will be considered in
the initial selection process, the total return, level and other
performance characteristics provided to an end user will only
relate to the desired Index. For example, if the user desires to
receive information about performance of the 30-year MBS Index,
total returns from all 15-year securities will be equal to "0" and
the displayed characteristics will only relate to 30-year
mortgage-backed securities.
[0046] With respect to step 18 of FIG. 4, one of the main
characteristics of each of the provided indices is its total
return. For limited periods of time, the total return of each index
may be defined as an average total return of the bonds represented
in that index. Hence, to calculate the total return over a short
period of time it is first necessary to calculate which bonds are
represented in the index during that time period, using the
algorithm described above, and second, to calculate the return of
each bond. For longer periods of time the total return is computed
by compounding over a set of disjoint and complete subintervals of
time.
[0047] As described above, on the last business day of each month a
set of relative weights for issues included in the Index is
introduced. For the month following the month when calculations
took place these weights represent relative proportions of each
pass-through selected to be in the Index. Therefore, from the last
business day of any month to any business day of the following
month, a fixed portfolio can represent the Index. Accordingly, over
this time period the Index total return is equal to the total
return of the representative portfolio. For longer time intervals,
representative portfolios may be used to calculate the
month-to-month returns. Compounding these returns gives the Index
return over several months.
[0048] For example, to calculate the Index total return from Oct.
14, 2000 until Dec. 15, 2000, representative portfolios may be used
to calculate the total returns for the months of Oct. 14, 2000 to
Oct. 31, 2000, of Oct. 31, 2000 to Nov. 30, 2000, and of Nov. 30,
2000 until Dec. 15, 2000. The product of these three numbers is the
Index total return from Oct. 14, 2000 to Dec. 15, 2000. The total
return from Oct. 14, 2000 until Oct. 31, 2000 is the total return
of the representative portfolio from Sep. 29, 2000 to Oct. 31, 2000
minus the return from Sep. 29, 2000 to Oct. 13, 2000.
[0049] To express the above example in algebraic form, consider a
date t.sub.1 in month k and a later date t.sub.2 in month n. When
k=n, the total return of the Index from the close of t.sub.1 until
the close of t.sub.2 is,
TR|.sub.t.sub.1.sup.t.sup.2=TR.sub.t.sub.2-TR.sub.t.sub.1 (Eq.
2.1),
[0050] where TR.sub.t.sub.1 is the month-to-date total return of
said index on the day t.sub.1, and TR.sub.t.sub.2 is the
moth-to-date total return of said index on the day t.sub.2.
[0051] Or when k=n-1,
TR|.sub.t.sub.1.sup.t.sup.2=(1+TR.sub.k-TR.sub.t.sub.1)(1+TR.sub.t.sub.2-
)-1 (Eq. 2.2),
[0052] where TR.sub.k is the total return of the Index for the
month k.
[0053] Or when k<n-1,
TR | t 1 t 2 = ( 1 + TR k - TR t 1 ) [ i = k + 1 n - 1 ( 1 + TR i )
] ( 1 + TR t 2 ) - 1 , ( Eq . 2.3 ) ##EQU00003##
where TR.sub.i is the total return of the Index for any
intermediate month between k and n In the preferred embodiment,
TR.sub.i is the total return of the Index from the close of the
last business day of the month i-1 to the close of the last
business day of the month i. Both TR.sub.i and TR.sub.t are defined
in Equations (3.1) and (3.2). Equations 2.1, 2.2 and 2.3 define the
operations to be performed in step 18 of the flow chart of FIG.
4.
[0054] From the end of one month to the end of the next month, the
total return of the Index is defined by the total return of a fixed
representative portfolio. To compute the return of a representative
portfolio one has to calculate the cost of buying the portfolio,
same-day settle on the earlier date; the gain from selling the
portfolio, same-day settle on the later date; and the value of any
pay downs earned.
[0055] Let the set {w.sub.i.sup.j}.sub.j=1.sup.n be the relative
weights of the securities composing the generic Index as of the
close of the last business day of month i-1. Then the total return
of the Index from the close of the last business day of the month
i-1 until the close of the last business day of month i is,
TR i = j = 1 n w i j p i i j TR i j j = 1 n w i j p i j , ( Eq .
3.1 ) ##EQU00004##
where TR.sub.t.sup.j is the total return of generic pass-through j
over this time interval as will be shown in Equation 4.1, discussed
with respect to step 16 of FIG. 4. The Equation 3.1 may be
rewritten for TR.sub.k as:
TR k = j = 1 n w k j p k j TR k j j = 1 n w k j p k j , ( Eq . 3.3
) ##EQU00005##
The price of each pass-through is preferably the same-day settle
price, measured on close of the last business day of month i-1 (or
k-1), and is denoted p.sub.i.sup.j (or p.sub.k.sup.j or simply
p.sub.i). Its computation is described further below (see Equations
5.1-5.5).
[0056] For any arbitrary business day t in any month i:
TR t = j = 1 n w i j p i j TR t j j = 1 n w i j p i j , ( Eq . 3.2
) ##EQU00006##
[0057] where TR.sub.t.sup.j is the total return of generic
pass-through j from the close of the last business day of the month
i-1 until the close of the business day, t.
[0058] As shown in FIG. 4, step 16, in order to calculate the total
return of the Index, the total return of each generic pass-through
has to be calculated first. In accordance with the preferred
embodiment of the present invention, formulas for calculating the
total return of each pass-through in the index for specific time
periods are provided. The total return calculations involve buying
and selling securities for the same-day settle price, however, the
only prices observable in the marketplace are for the standard PSA
(i.e., TBA) settle prices. Consequently the standard formulas
typically used in the industry are based on PSA settle prices. In
contrast, the method of calculating the total return of each
pass-through in accordance with the present invention uses formulas
which are based on the same-day settle prices, not standard PSA
settle prices. Therefore, a conversion algorithm is provided by the
present invention, as more particularly described in connection
with Equations 5.1-5.5.
[0059] Consider a mortgage-backed security of coupon c. At the
close of the last business day of a month an investor purchases $1
of the security for price p.sub.1. On the close of an arbitrary
business day of the following month the investor sells the
remaining principal for price p.sub.2. The investor receives the
coupon payment,
c 12 , ##EQU00007##
and the pay down (1-f) due the month of sale, which is invested at
rate r. These numbers suffice to calculate the total return of this
security.
[0060] For a generic pass-through j, of fixed-rate coupon c, the
total return from the close of business on the last business day
t.sub.1 of one month, to the close of business on an arbitrary
business day, t.sub.2, in the next month is defined as:
TR t 2 j = - p t 1 + f t 1 p t 2 + [ ( 1 - f t 1 ) + c 12 ] [ 1 + r
t 1 d 360 ] - k p t 1 ( Eq . 4.1 ) ##EQU00008##
[0061] The terms on the right hand side are:
[0062] p.sub.t=price of the security on the close of t, same-day
settle.
[0063] r.sub.t=1-month BBA LIBOR on the close of t (provided by
Reuters).
[0064] f.sub.t=the pay-down factor of the pass-through as best
determined by day t.
d = { 25 - day of the month of t 2 , if bond FNMA 15 - day of the
month of t 2 , if bond GNMA or FHLMC k = { + 1 , if d < 0 - 1 ,
if d < 0 ##EQU00009##
[0065] Data for all of these terms is preferably input into the
system.
[0066] The algorithm for computing f.sub.t is provided in Equation
4.3. Prices, rates, and coupons are preferably all considered as
decimals, not percents. If t.sub.2 is the last business day of
month i, it can be written:
TR.sub.i.sup.j=TR.sub.t.sub.2.sup.j (Eq. 4.2)
[0067] For a specific mortgage pool, the pay down for month i is
preferably determined by the actual loans represented in the pool
during month i-1. Therefore, two separate pools with the same
characteristics can have different total returns.
[0068] In the preferred embodiment of the present invention, the
Index represents generic pass-throughs, that is, pass-throughs
representative of all pools of the same coupon, and original term,
issued by the same agency. The total return of a generic
pass-through is a weighted average of all of the pass-throughs it
represents.
[0069] For month i, the factor of the generic pass-through with
coupon c, original term t, issued by agency a may be determined by
the following algorithm. First the working set is defined as all
MBS pools issued by the agency a, of original term t, with fixed
coupon c as of the first of month i. With this set, the maximum
WALA (Weighted Average Loan Age) for TBA pools will be defined
as:
n cutoff = min { n : principle of pools with WALA .ltoreq. n
principle of all pools .gtoreq. 2.5 % } , ##EQU00010##
[0070] with all numbers as of the first of month i+1, then the
subset of TBA pools will be
A={pools: WALA.ltoreq.n.sub.cutoff as of the first of the month
i+1}.
[0071] Using this set the pay-down factor of month i is calculated
as:
f i = .alpha. .di-elect cons. .LAMBDA. .rho. .alpha. , i - .alpha.
.di-elect cons. .LAMBDA. .rho. .alpha. , i + 1 .alpha. .di-elect
cons. .LAMBDA. .rho. .alpha. , i ( Eq . 4.3 ) ##EQU00011##
[0072] with .rho..sub..alpha.,i as the principal outstanding of
pool .alpha. as of the first of month i. If the WALA of a pool is
unknown, it is estimated using CAGE. f.sub.i is the generic
pay-down factor for the month i. Equation 4.3 creates a sequence of
monthly pay-down factors for each generic pass-through. For a
specific date t in month i, the pay-down factor of a generic
pass-through as best known by t is preferably denoted f.sub.t. The
value f.sub.t is preferably the latest factor that would be known
as of the close of t. For example if f.sub.i is known by t, then
f.sub.t is defined as f.sub.i. Or else, if f.sub.i-1 is known by t,
then the f.sub.t is defined to be f.sub.i-1. Alternatively, the
f.sub.t will be defined as f.sub.i-2. The advantage of the above
definition of f.sub.t is that it makes the Index independent of
subjective prepayment models. Data related to each term found in
Equation 4.3 is preferably input into the system so that the above
calculations may be performed.
[0073] As described above, the provided formulas for calculating
the total return of the Index is based on the same-day settle
price, not standard PSA settle price. As shown for example in FIG.
3, the Bond Market Association announces the MBS settlement dates
on which all purchasing transactions have to be settled. However,
if the formula for calculating the total return of each generic
pass-through is based on this TBA (or PSA) price, the calculation
will necessarily include some element of estimating future cash
flows. Consequently, in accordance with the preferred embodiment of
the present invention, the formulas for converting from standard
PSA settle prices to same-day settle prices are provided. As
explained above, standard PSA settle price is the market observed
price. Equations 5.1-5.5 provide formulas for converting the
standard PSA settle price, 1-month forward, to the same-day settle
price. Data related to each term found in Equations 5.1-5.5 is
preferably input into the system so that the above calculations may
be performed.
[0074] There are two major differences between quoting prices with
standard PSA settle and with same-day settle. First, standard PSA
settle assumes that, while the price is agreed upon today, no
payment is made until sometime in the future. Same-day settle
requires payment today. Therefore a PSA settle price includes some
time value that must be discounted to convert to today's dollars.
Second, pass-throughs begin paying principal and interest to the
bondholder the month following settlement. Therefore, purchasing
securities 1-month forward standard PSA settle entitles the buyer
to the pay down two months after the purchase day and not the pay
down one month after the purchase date. For same-day settle the
buyer is entitled to both of these pay downs.
[0075] Let {tilde over (p)} represent the standard PSA settle price
for a given TBA pass-through on a given date. The dirty price of
that security (actual number of dollars expected for $1 of
principal) is,
p ~ + c 12 d 1 360 ( Eq . 5.1 ) ##EQU00012##
where c represents the coupon of the pass-through expressed as a
decimal (e.g. 7% implies c=0.07), d.sub.1 represents the number of
days into the month that 1-month forward standard PSA settle
occurs. For example, if the standard PSA settlement for June 2000
is the 13th, then for any date in May 2000, d.sub.1=12.
[0076] Discounting this price to the day in question, the following
formula is derived:
p ~ + c 12 d 1 360 1 + r d 2 360 , ( Eq . 5.2 ) ##EQU00013##
where r represents the discount funding rate; and d.sub.2
represents the number of days in between the purchase date and the
standard PSA settle date 1-month forward inclusive of the former
and exclusive of the latter. The rate is quoted as a decimal, e.g.
a rate of 6.5 percent is written as 0.065.
[0077] The value of the pay down and interest for the month
following the purchases date is the sum of the two payments
(discounted),
( 1 - f t ) + c 12 1 + r d 3 360 , ( Eq . 5.3 ) ##EQU00014##
[0078] with f representing the factor (derived from Equation 4.3)
for that bond for that month. In the above equation, d.sub.3
represents the number of days between the purchase date and the
25th of the next month (for FNMA MBS) or the 15th of the next month
(for GNMA or FHLMC MBS), inclusive of the former and exclusive of
the latter.
[0079] Assuming business day t falls in month i, the same-day
settle price for a pass-through at the close of business on t,
denoted by p.sub.t, can be calculated by combining (5.2) and (5.3)
((5.2) is modified because (5.3) represents some of the principal).
The following equation is then derived:
p t = p ~ t + c 12 d 1 30 1 + r d 2 360 f t + ( 1 - f t ) + c 12 1
+ r d 3 360 , ( Eq . 5.4 ) ##EQU00015##
[0080] with {tilde over (p)}.sub.t being the TBA price of a
pass-through 1-month forward standard PSA settle on the close of
business on t, as quoted for example by the CSFB Pass-Through desk.
The pay-down factor of the pass-through as best determined by t,
f.sub.t, is selected from the sequence of monthly pay-down factors
derived from the Equation 4.3. In this preferred embodiment, r is a
1-month BBA LIBOR (British Bankers Association London Interbank
Offered Rate) at the close of date t, if t is the last business day
of the month. If not, then r is 1-month BBA LIBOR at the close of
the last business day of month i-1. d.sub.1 is the number of days
between the first of month i+1 and the standard PSA settlement date
in month i+1 inclusive of the later and exclusive of the former.
The number of days between t and the standard PSA settlement date
of month i+1 inclusive of the former and exclusive of the latter is
shown in the (5.4) as d.sub.2. For FNMA pass-throughs, d.sub.3 is
the number of days between the 25th of month i+1 and t inclusive of
the former and exclusive of the latter. For GNMA and FHLMC
pass-throughs, d.sub.3 is the number of days between the 15th of
the month i+1 and date t, inclusive of the former and exclusive of
the latter. The values {tilde over (p)}.sub.t, r.sub.t, and c are
often quoted as percents, however, in (5.4) they are all decimals.
The source of 1-month BBA LIBOR is Reuters, and is defined in
BRITISH BANKERS' ASSOCIATION (2000) Libor Official Definition
(incorporated herein by reference). The BBA LIBOR and its
definition may also be found at www.BBA.ORG.UK. The source of the
PSA settlement dates is The Bond Market Association and may be
obtained from www.bondmarket.com, as shown for example in FIG.
3.
[0081] When t is the last business day of the month i-1:
p.sub.t=p.sub.i, Eq. 5.5
[0082] In accordance with the preferred embodiment of the present
invention, each index may be characterized by its level. The level
of the Index is defined by Equations 6.2 and 6.3 and is calculated
in step 20 in FIG. 4. It should be noted that level determination
is not a required step of the present invention, and any other tool
known in the art may be used with the provided MBS Index to
evaluate its performance.
[0083] In accordance with the preferred embodiment, the starting
date of the index is input into the system. Equation 6.2 sets a
starting level for this chosen starting date. Equation 6.3 sets the
percent level change between two days to be the total return of the
Index between those two days. The total return of the Index is
determined from the Equations 2.1, 2.2 or 2.3 above.
[0084] The level is a useful tool for evaluating an investment in
the Index. Assuming monthly reinvestment in the Index, a cash
investment of $X on the date t.sub.1 will show either a profit or a
loss by a latter date t.sub.2 such that
$ X P t 2 - P t 1 T t 1 , ( Eq . 6.1 ) ##EQU00016##
where P.sub.t is the level of the Index on day t. Of course, the
investor's true total return will depend on how the monthly
payments are reinvested.
[0085] If on the close of 12/31/93 the initial level of the Index
is defined as:
P.sub.12/31/93=100 (Eq. 6.2),
then on the close of any subsequent date t the level will be
P t P 12 / 31 / 93 = 1 + TR 12 / 31 / 93 t , ( Eq . 6.3 )
##EQU00017##
where TR|.sub.12/31/93.sup.t is the total return of the Index from
the close of 12/31/93 until the close of t. Over time movements in
the levels are determined by the total return of the indices. Of
course, the date 12/31/93 is selected arbitrarily and may be
substituted by the actual start date of the index.
[0086] In accordance with the preferred embodiment of the present
invention, a system 50 for generating and managing an MBS Index is
provided to perform the above described method. FIG. 2 is a
functional block diagram of an embodiment of the invention showing
discrete components and information flow within the system 50. To
accomplish the described steps of the algorithm 10, system 50
preferably comprises a market data input device 26 (which may
include an input/output interface 2 and input device 3, as shown in
FIG. 1) for imputing necessary data related to outstanding
mortgage-backed securities. Such data will include original term,
coupon value, issuing agency, and outstanding principal on each
outstanding mortgage-backed security. Additional data inputs into
the system 50 through the market data input 26 preferably include
TBA settle prices, PSA settle dates and BBA LIBOR rates.
[0087] The input data received by the system from the input 26 is
preferably classified in the classification processor 28. The
classification processor 28, which may be part of the CPU 1 shown
in FIG. 1, classifies all outstanding mortgage-backed securities in
accordance with their original term, coupon value, and issuing
agency. Classified securities are then aggregated into pools in
accordance with their coupon and original term. The data and
composition of all aggregated pools are then outputted from the
classification processor into the central hub 24. As will be
obvious to one of ordinary skill in the art, it is not necessary to
include the central hub 24 into the system 50, and all described
data transfers may be accomplished by directly connecting the
discrete components of the system to each other. Input processors
28-42 of FIG. 2 may all be included in the single CPU 1 of FIG.
1.
[0088] The central hub 24 is preferably connected to an Index
database 44 for storing data related to the provided MBS Index. The
Index database 44 may also store some tentative data, for example
results of intra-step calculations, for future use. Although shown
as a single component in FIG. 2, the Index database may comprise
several components, for example, RAM 5, shown in FIG. 1, a hard
drive, a connected zip drive or a separate database server. Output
of every discrete component of the system 50 is preferably input
into the central hub 24 and is further transmitted into the Index
database 44. Selected data stored in the Index database may be
outputted to the output terminal 46. The output terminal may
include or be connected to the input/output interface 6 and the
output device 7 as shown in FIG. 1.
[0089] On the last business day of month i, the composition of all
aggregated pools and other data related to these pools is input
into an Index composition processor 30 from the central hub 24. The
logic flow chart of the processing logic for the index composition
determination is shown in FIG. 5. Logic conceptually begins at
block 301 and proceeds to block 302 where the aggregated pools and
their associated data are entered into the composition processor.
At block 303, based on the input data, a total outstanding
principal for all aggregated pools is calculated. One of the
aggregated pools is selected for determination in block 304, and
its outstanding principal is calculated in block 305. The results
of calculations of blocks 303 and 305 are input into block 306,
where the inclusion criterion is determined in accordance with
Equation 1.2. At logic block 307, the processor will determine the
original term of the pool selected in 304 and, if the original term
is 30-years (360 months), the inclusion criterion x.sub.c,t is
compared to 1.5% at logic block 308. If however, the original term
is 15 years (180 months) the inclusion criterion is compared to
0.4% at logic block 309. Aggregated pools having a 30-year original
term are included in the Index at block 311 if the expression at
block 308 is determined to be true. Aggregated pools having a
15-year original term are included in the Index at block 311 if the
expression at block 309 is determined to be true. Otherwise, the
pool is discarded at block 310. After either the inclusion or
exclusion of the pool selected in 304, the process is repeated from
block 304 until all aggregated pools are considered. The Index
composition, i.e., a list of all selected securities, is outputted
into the central hub 24 and stored in the Index database 44,
preferably with a month for which this Index was generated. As
explained above, the composition of the Index will remain constant
for one month. On the last business day of the next month the index
composition processor is activated again and the selection process
repeats.
[0090] Relative weights of each mortgage-backed security included
in the Index are preferably calculated in the weights processor 32,
FIG. 2. The logic flow chart of the processing logic for the
weights determination is shown in FIG. 6. Logic conceptually begins
at block 321 and proceeds to input block 322 where the inclusion
criteria from block 306 and total outstanding principal on each
selected security from market data input 26 of FIG. 2 are entered
into the weights processor. Total outstanding principal on all
securities included in the Index is calculated at block 323, and
the relative weight of each security is determined at block 324
using the Equation 1.3. The relative weights are outputted from the
weights processor 32, input into the central hub 24 and preferably
stored in the Index database 44.
[0091] In the preferred embodiment, the system 50 is also provided
with a date calculator and internal calendar 38. As described above
in connection with steps 16 and 18 of the algorithm 10, shown in
FIG. 4, various dates and differences between them are used to
calculate factor f.sub.t, total return of each pass-through, total
return of the Index, same-day-settle price and level of the Index.
As would be obvious to one of ordinary skill in the art, these
dates may be observed in the market and if the differences between
them are required for further calculations, these differences can
be determined within the individual processors, where these further
calculations are to be performed. However, for purposes of
efficiency it is preferred that the date on which each calculation
is performed (also referred to in this disclosure as "the current
date" and "t.sub.2") is determined by the provided internal
calendar. It is also preferred that all manipulations and
calculations involving dates are made in the provided date
calculator associated with the internal calendar. For example,
instead of calculating the term d twice, as this term is required
in the calculations of total return of each individual pass-through
and same-day-settle price (in block 34 and block 42 of FIG. 2), it
can be calculated once in the date calculator 38 and then input
into blocks 34 and 42 when necessary.
[0092] Pay-down factor f.sub.i for each month is calculated in the
factor processor 40 of FIG. 2. The logic flow chart of the
processing logic for the factor determination is shown in FIG. 7.
Logic conceptually begins at block 401 and proceeds to input block
402 where the total outstanding principal on each selected security
is entered into the factor processor. The outstanding principal
data may be input at 26 of FIG. 2 or may already be stored in the
Index database 44 during performance of prior operations (e.g.
steps 12 and 14 of FIG. 4). Principal outstanding of all securities
within pool .alpha. for month i is calculated at block 403, and
principal outstanding of all securities within pool .alpha. for
month i+1 is calculated at block 404. These outstanding principals
for two consecutive months are then input into the factor
calculator 405, where the pay-down factor is determined in
accordance with Equation 4.3. The pay-down factor f.sub.i is
outputted from the factor processor 40, input into the central hub
24 and stored in the Index database 44.
[0093] Same-day-settle price for each security is calculated in the
same-day-settle price processor 34 of FIG. 2. The logic flow chart
of the processing logic for the determination of these prices is
shown in FIG. 8. Logic conceptually begins at block 341 and
proceeds to input block 342 where the 1-month BBA LIBOR, TBA price,
and coupon value of each security included in the Index is entered
into the same-day-settle price processor from the market data input
26 of FIG. 2. Additionally, date differences d.sub.1, d.sub.2 and
d.sub.3 are entered into the input block 342 from the date
calculator 38. In block 343 the latest known pay-down factor
f.sub.t is selected from the Index database 44 where all pay-down
factors are stored. Using the input data, the same-day-settle price
calculator 34 determines this price in accordance with Equation 5.4
in block 344. The same-day-settle price is outputted from its
processor 34, input into the central hub 24 and stored in the Index
database 44. The above described algorithm is repeated for every
security included into the Index in the month of calculations.
[0094] Total return of the Index from the date t.sub.1 of month k
to date t.sub.2 of month n is calculated in the total return
processor 42 of FIG. 2. The logic flow chart of the processing
logic for the total return determination is shown in FIG. 9. Logic
conceptually begins at block 421 and proceeds to input block 422
where all necessary dates and differences between them are entered
from block 38, coupon values and BBA LIBOR rates are entered from
input 26 (or from the Index database 44 if these rates were
previously stored there), same-day-settle prices are either entered
from block 34 or retrieved from the database 44, the latest known
pay-down factor is retrieved from the database 44, and the relative
weights are either entered from block 32 or retrieved from the
database 44. Total return of each individual pass-through in the
Index is calculated at block 423 in accordance with Equation 4.1.
Total returns of the Index for individual months (TR.sub.i) are
either calculated at block 424 in accordance with Equation 3.1 or
retrieved from the database 44 if available, and total returns of
the Index from the first of the month until a desired date (e.g.
t.sub.1 or t.sub.2) within that month are calculated at block 425
in accordance with Equation 3.2. These calculated total returns are
then input into the logic block 426. Total returns for individual
months are preferably also outputted for storage in the database 44
so that they can be later retrieved from the database without
repeating the above calculations. At logic block 426, the processor
will determine the difference between the months k and n. If k and
n are the same month (i.e., k=n), the total return of the index
between t.sub.1 and t.sub.2 is calculated at block 427 according to
Equation 2.1. If k and n are not the same, the system will proceed
to logic block 428 where the system will determine whether k and n
are consecutive months (i.e., k=n-1). If k and n are in fact
consecutive, the processor will calculate the total return of the
Index at block 429 using Equation 2.2. Alternatively, the processor
will calculate the total return of the Index at block 430 using
Equation 2.3. The output of blocks 427, 429 and 430 is preferably
input into the database 44 for storage. Additionally, in accordance
with the preferred embodiment, a month-to-day total return of the
Index is outputted onto the output terminal 46 as a benchmark
measuring the performance of the provided MBS Index.
[0095] System 50 is further provided with a level processor 36
where the level of the Index is determined in accordance with
Equation 6.3. The starting level of the Index is preferably
arbitrarily assigned, stored in the database 44 and retrieved for
calculations by the level processor 36 when necessary. The total
return for the life of the index may be calculated in the total
return processor 42 and input into the level processor 36. The
resulting level of the Index is input into the central hub 24 and
stored in the Index database 44. Additionally, in accordance with
the preferred embodiment, a daily level of the Index is outputted
onto the output terminal 46 as a benchmark measuring the
performance of the provided MBS Index. Any other variable stored in
the Index database 44 may also be displayed on the output terminal
46 if desired.
[0096] MBS Indices provided in accordance with the present
invention may be used as a benchmark for objectively and accurately
measuring the performance of the MBS sector of the market. The
indices will preferably be published, or otherwise made available,
on a monthly basis. The method provided by the present invention
for generating MBS indices may be used by portfolio managers for
creating index funds mirroring or reflecting the provided MBS
indices. The algorithm may be encoded into a computer program and
distributed to users on a CD-ROM or another readable storage
device. Alternatively, this MBS index-generating program may be
stored on a web server or an enterprise server and streamed,
downloaded or otherwise provided to interested users on an
as-needed basis. Since the composition of each index is preferably
held constant throughout a month, fund managers avoid the need to
match the provided benchmark's moves on a daily basis.
Additionally, the provided system and method allow users to easily
and automatically rebalance the index at or near the end of the
month.
[0097] Having described this invention with regard to specific
embodiments, it is to be understood that the description is not
meant as a limitation since further variations or modifications may
be apparent or may suggest themselves to those skilled in the art.
For example, the provided method may easily be modified to generate
other types of MBS indices. It is intended that the present
application cover such variations and modifications as fall within
the scope of the appended claims.
* * * * *
References