U.S. patent application number 14/268279 was filed with the patent office on 2014-08-28 for self-perpetuation of a stochastically varying resource pool.
This patent application is currently assigned to FMR LLC. The applicant listed for this patent is FMR LLC. Invention is credited to Andrew Dierdorf, Stephen D. Fisher, Jurrien Timmer, Walter Tsui.
Application Number | 20140245320 14/268279 |
Document ID | / |
Family ID | 40089207 |
Filed Date | 2014-08-28 |
United States Patent
Application |
20140245320 |
Kind Code |
A1 |
Timmer; Jurrien ; et
al. |
August 28, 2014 |
Self-Perpetuation of a Stochastically Varying Resource Pool
Abstract
A computer-readable medium has encoded thereon software for
maintaining a steady-state worth of an inhomogenous renewable
resource pool. The software includes instructions for causing a
data-processing system to evaluate an indicator of a historical
worth of the resource pool, to determine a draw amount at least in
part on the basis of this indicator, and to output data
representative of that draw amount.
Inventors: |
Timmer; Jurrien;
(Manchester, MA) ; Dierdorf; Andrew; (Watertown,
MA) ; Tsui; Walter; (Wellesley, MA) ; Fisher;
Stephen D.; (Belmont, MA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
FMR LLC |
Boston |
MA |
US |
|
|
Assignee: |
FMR LLC
Boston
MA
|
Family ID: |
40089207 |
Appl. No.: |
14/268279 |
Filed: |
May 2, 2014 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11755426 |
May 30, 2007 |
8756128 |
|
|
14268279 |
|
|
|
|
Current U.S.
Class: |
718/104 |
Current CPC
Class: |
G06Q 40/06 20130101;
G06F 9/50 20130101 |
Class at
Publication: |
718/104 |
International
Class: |
G06F 9/50 20060101
G06F009/50 |
Claims
1. A computer-readable medium having encoded thereon software for
maintaining a steady-state worth of an inhomogenous renewable
resource pool, the software including instructions for causing a
data-processing system to: evaluate an indicator of a historical
worth of the resource pool; at least in part on the basis of the
indicator, determine a draw amount; and output data representative
of the draw amount.
2. The computer-readable medium of claim 1, wherein the
instructions for causing a data processing system to evaluate an
indicator comprise instructions for causing a data processing
system to retrieve time series data indicative of historical worth
of the resource pool.
3. The computer-readable medium of claim 2, wherein the
instructions for causing a data processing system to evaluate an
indicator comprise instructions for causing a data processing
system to low-pass filter the time series data.
4. The computer-readable medium of claim 2, wherein the
instructions for causing a data processing system to evaluate an
indicator comprise instructions for causing a data processing
system to determine an average value of the time series data.
5. The computer-readable medium of claim 1, wherein the
instructions for causing a data processing system to evaluate an
indicator comprise instructions for causing a data processing
system to retrieve synthetic time series data indicative of what a
historical worth of the renewal resource pool would have been.
6. The computer-readable medium of claim 1, wherein the
instructions for causing a data processing system to determine a
draw amount comprise instructions for causing a data processing
system to determine a pre-determined fraction of the historical
worth.
7. The computer-readable medium of claim 1, wherein the software
further comprises instructions for causing a data processing system
to adjust the draw amount on the basis of an external constraint on
the draw amount.
8. The computer-readable medium of claim 1, wherein the software
further comprises instructions for retrieving data indicative of
historical worth of an endowment fund, and wherein the instructions
for causing a data processing system to evaluate an indicator of
historical worth comprise instructions for causing a data
processing system to retrieve a plurality of selected values
indicative of a corresponding plurality of worths of the endowment
fund at each of a plurality of times.
9. A method for maintaining a steady-state worth of an inhomogenous
renewable resource pool, the method comprising: evaluating an
indicator of a historical worth of the resource pool, at least in
part on the basis of the indicator, determining a draw amount; and
outputting data representative of the draw amount.
10. The method of claim 9, wherein evaluating an indicator
comprises retrieving time series data indicative of historical
worth.
11. The method of claim 10, wherein evaluating an indicator
comprises low-pass filtering the time series data.
12. The method of claim 11, wherein evaluating an indicator
comprises determining an average value of the time series data.
13. The method of claim 9, wherein evaluating an indicator
comprises retrieving synthetic time series data indicative of what
a historical worth of the renewal resource pool would have
been.
14. The method of claim 9, wherein determining a draw amount
comprises determining a pre-determined fraction of the historical
worth.
15. The method of claim 9, further comprising adjusting the draw
amount on the basis of an external constraint on the draw
amount.
16. The method of claim 9, further comprising retrieving data
indicative of historical worth of an endowment fund, and wherein
evaluating an indicator of historical worth comprises retrieving a
plurality of selected values indicative of a corresponding
plurality of worths of the endowment fund at each of a plurality of
times.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application is a continuation of and claims priority to
U.S. application Ser. No. 11/755,426, filed on May 30, 2007,
entitled "Self-Perpetuation of a Stochastically Varying Resource
Pool," the entire contents of which are hereby incorporated by
reference.
FIELD OF DISCLOSURE
[0002] The disclosure relates to control systems, and in particular
to feedback control systems for management of a resource pool.
BACKGROUND
[0003] A control system manipulates certain variables, referred to
as "control variables," associated with a controlled system. It
does so in an attempt to cause the controlled system to achieve a
particular goal, often referred to as a "set point." A feedback
control system receives information indicative of an error between
the actual performance of the controlled system and the set point.
It then uses this information to manipulate the control variable in
an effort to reduce this error.
[0004] One system that can be placed under control of a feedback
control system is a resource pool. Resource pools that include
different kinds of resources, are referred to as inhomogeneous
resource pools. These resource pools have a mixture of resources
that fluctuate in their amounts and/or values. These fluctuations
are often random fluctuations. Because of the random nature of
fluctuation, such resource pools are referred to as "stochastically
varying" resource pools.
[0005] Resources in resource pools can be of various types,
including non-renewable resources, such as fossil fuels, or
renewable resources, such as timber. Resources can also include
obligations or ownership interests, the values of which depend on
supply and demand of underlying resources, or of goods and services
that are dependent on such resources. Such obligations and
ownership interests are often referred to as bonds or shares
associated with some entity.
SUMMARY
[0006] In one aspect, the invention features a method for
maintaining a steady-state worth of an inhomogeneous renewable
resource pool. Such a method includes evaluating an indicator of a
historical worth of the resource pool, at least in part on the
basis of the indicator, determining a draw amount; and outputting
data representative of the draw amount.
[0007] Practices include those in which evaluating an indicator
includes retrieving time series data indicative of historical
worth, those in which evaluating an indicator includes low-pass
filtering the time series data, those in which evaluating an
indicator includes determining an average value of the time series
data, and those in which evaluating an indicator includes
retrieving synthetic time series data indicative of what a
historical worth of the renewal resource pool would have been.
[0008] In some practices, determining a draw amount includes
determining a pre-determined fraction of the historical worth.
[0009] Other practices include adjusting the draw amount on the
basis of an external constraint on the draw amount.
[0010] Yet other practices include retrieving data indicative of
historical worth of an endowment fund, and wherein evaluating an
indicator of historical worth includes retrieving a plurality of
selected values indicative of a corresponding plurality of worths
of the endowment fund at each of a plurality of times.
[0011] In another aspect, the invention features a
computer-readable medium having encoded thereon software for
causing a data-processing system to carry out any of the foregoing
methods. Such a data-processing system can be a distributed system
in which constituent processors, including possibly remote
processors, cooperate to carry out the foregoing methods.
[0012] A system as described herein can be implemented in hardware
or software, or a combination of both. The system can be
implemented in computer programs using standard programming
techniques following the method steps and figures described herein.
Suitable programs include those designed to execute on programmable
computers and/or data processing systems, each comprising a
processor, a data storage system (including memory and/or storage
elements), at least one input device, and at least one output
device, such as a display device or printer. Program code is
applied to input data to perform the functions described herein and
generate output information. The output information is applied to
one or more output devices such as a display device, as described
herein.
[0013] Each program is preferably implemented in a high level
procedural or object oriented programming language to communicate
with a computer system. However, the programs can be implemented in
assembly or machine language, if desired. In any case, the language
can be a compiled or interpreted language.
[0014] Each such computer program is preferably stored on a storage
medium or device (e.g., ROM or magnetic diskette) readable by a
general or special purpose programmable computer, for configuring
and operating the computer when the storage media or device is read
by the computer to perform the procedures described herein. The
system can also be considered to be implemented as a
computer-readable storage medium, configured with a computer
program, where the storage medium so configured causes a computer
to operate in a specific and predefined manner to perform the
functions described herein.
[0015] These and other features of the invention will be apparent
from the following detailed description and the accompanying
figures, in which:
BRIEF DESCRIPTION OF THE FIGS.
[0016] FIG. 1 shows a control system for of a controlling a
renewable inhomogeneous resource pool;
[0017] FIG. 2 is a block diagram showing the operation of the
control system of FIG. 1; and
[0018] FIG. 3 is a block diagram showing the operation of the
control system subject to an external constraint on depletion
amount.
DETAILED DESCRIPTION
[0019] Referring to FIG. 1, an inhomogeneous resource pool 12 is
one that includes resources of different types. For example, the
resources in a resource pool 12 may be different types of fuel,
such as coal and oil. Or the resources may be different types of
grain to be used in connection with manufacture of foods, or
different raw materials to be used in a manufacturing process.
[0020] A resource pool 12 may contain assets, such as different
types of financial instruments, whose individual values fluctuate.
In some cases, a resource pool 12 may include, among its
constituent assets, other resource pools, such as mutual fund
shares. A resource pool 12 of the type described herein can also
include, as a constituent asset thereof, one or more resource pools
of the type described herein. A resource pool 12 can be
characterized by the type of resource and the quantity of that
resource. For example, a resource pool 12 can be an oil deposit
having a certain number of barrels of oil reserves.
[0021] This characterization provides a basis for estimating the
worth of that resource pool 12. However, in the case of an
inhomogenous resource pool, this exercise becomes problematic. For
example, if the resource pool 12 includes a gold mine, an oil
field, a set of shares, and a set of bonds, it can no longer be
characterized by a type and quantity of a resource.
[0022] To avoid such difficulties, one often characterizes an
inhomogenous resource pool 12 by estimating a scalar value
indicative of the worth of each constituent resource. These scalar
values can then be added together to provide an overall worth for
the entire resource pool 12 itself
[0023] Another way to assign a worth to a resource pool 12 is to
assign values to entities that depend in some way on underlying
resources. Typical entities include companies organized to, in some
way, profit from the existence of one or more resources. The worths
of such companies are made manifest by worths of associated
financial instruments and derivatives thereof. As a result, the
worth of an inhomogenous resource pool 12 can be measured by worths
of assets, such as different types of financial instruments.
[0024] In some cases, the resources in a resource pool 12 are
renewable resources. For example, a resource may be a stand of
timber, a herd of livestock, or a colony of microbes to be used,
for example, in vaccine production, or as leavening or fermentation
agents in food production. However, renewability does not
necessarily mean immortality. Even a renewable resource pool can be
completely depleted. For example, if the harvest rate for a stand
of timber exceeds the growth rate, or replacement rate, eventually
that stand will be fully depleted.
[0025] A resource pool 12 that includes financial instruments can
likewise be fully depleted if the withdrawal, or distribution rate,
exceeds the rate at which those instruments appreciate in value. In
many cases, it is desirable to control a renewable resource pool
such that the renewal rate of the pool will match a rate at which
resources are drawn from that resource pool 12, hereafter referred
to as the "draw rate." A renewable resource pool 12 controlled in
this way will tend to achieve a steady-state worth. This
steady-state worth shall be referred to herein as the "capital" of
the resource pool 12.
[0026] The aggregate worth of resources within the resource pool 12
depends, to some extent, on a fluctuation signal 14 provided by a
stochastic disturbance source 16. For example, in the case of an
inhomogeneous resource pool 12 containing different stands of
timber, there may be fluctuations in the growth rate and harvest
rate of each stand. Or, when the resource pool 12 is a collection
of one or more mutual funds, or financial instruments, the
stochastic disturbance source 16 may be viewed as representing the
fluctuations in the underlying supply and demand associated with
those resources. These fluctuations in supply and demand, as
represented by the fluctuation signal 14, cause the value of the
resources in the resource pool 12 to vary in a way that forecloses
complete predictability. The stochastic nature of the fluctuation
signal 14 causes the resource pool 12 to be a
stochastically-varying resource pool 12.
[0027] In many cases, one may wish to maintain a resource pool 12
at some steady-state condition. Such a steady-state condition can
be characterized by a constant quantity of the various resources
that make up the resource pool 12. Alternatively, the steady-state
condition can be characterized by a constant, or essentially
constant capital, typically measured in units of a normalized
currency. As used herein, a normalized currency is one that is
normalized to its value at a particular time. An example of a
normalized currency is an inflation or deflation adjusted dollar,
which is pegged to the value that it attains at a particular
time.
[0028] For example in managing an endowment, one often wishes to
maintain the worth of the endowment's capital, while simultaneously
providing a worth stream to fund whatever is to be endowed. In such
cases, this output worth stream tends to draw worth from the
capital. By adjusting the draw rate to match the capital's renewal
rate, one can maintain the capital at a constant, or nearly
constant, worth.
[0029] The fluctuation signal 14 often includes high-frequency
components that make it difficult to maintain the capital of the
resource pool 12. For example, a short-term fluctuation that
results in an unexpected worth decrease might result in an
unnecessarily decreased draw rate. Conversely, a fluctuation that
results in an unexpected worth increase might result in an
excessively spendthrift draw on the resource followed by an
excessively parsimonious draw rate to compensate for the earlier
profligacy.
[0030] In response to the fluctuation signal 14, a controller 18
urges the resource pool 12 to maintain a steady-state value. It
does so by controlling either or both: (1) the constituent
resources of the resource pool 12; and (2) the draw rate of the
resource pool 12. It controls the former by generating a resource
control signal 20, and it controls the latter by generating a draw
control signal 22.
[0031] A feedback control signal 23 includes information indicative
of the state of the resource pool 12 as well as information
representative of economic indicators. In part on the basis of the
feedback signal 23, the controller 18 generates a resource control
signal 20 that dynamically controls the underlying contents of the
resource pool 12. For example, in a financial services application,
the resource control signal 20 may cause re-allocation between
volatile and non-volatile assets, such as between stocks and bonds,
or between different classes of stocks (e.g. small-cap and
large-cap stocks, common and preferred stocks, and/or domestic and
international stocks), between individual stocks, or it may
reallocate resources into or out of real estate based instruments,
commodity-based instrument, or hedge funds.
[0032] The draw control signal 22 acts as a regulator to govern the
rate at which resources are drawn from the resource pool 12. This
draw control signal 22 depends in part on the passage of a time
series through a low-pass digital filter. Preferably, the time
series represents a historical worth of the resource pool 12. For
example, in some practices, the controller 18 inspects a database
containing time series data representative of historical values
indicative of the worth of the resource pool 12. In some cases, the
values of the time series are calculated assuming that none of the
draw amounts are re-invested but that any capital gains or
dividends in excess of the draw amount are re-invested.
[0033] A typical time series includes monthly samples of the worth
of the resource pool 12 over some evaluation interval. A suitable
evaluation interval, for example, could include thirty-six monthly
samples extending back from the present, with the first sample
being obtained during the preceding month. However, the extent of
the evaluation interval, the number of samples within it, and the
interval between those samples can be varied to suit the particular
application. The passage of such a time-series through a low-pass
digital filter tends to filter out high-frequency components of the
stochastically varying fluctuation signal 14. The filter transfer
function is controlled by varying the evaluation interval. In the
present embodiment, the filter transfer function would drop to zero
at some frequency, the selection of which sets the filter's
bandwidth. The resulting discontinuity in the filter transfer
function would correspond to ignoring time series data beyond the
edge of the evaluation interval. However, one could also provide a
filter-transfer function that gradually declines to zero. This
would correspond to weighting the time-series data, and would
provide a way to magnify the influence of some portions of the data
without completely ignoring other portions of the data.
[0034] The output of the digital filter can be viewed as an average
worth of the resource pool 12 during the evaluation interval after
a draw for that month has occurred. This results in a moving
average of the worth of the resource pool 12. In response to this
output, the controller 18 generates a draw control signal 22 that
causes a pre-determined fraction of the worth of the resource pool
12 to be drawn. In some practices, the pre-determined fraction is
5%, however other fractions can also be used. The controller 18
selects both the draw control signal 22 and the resource control
signal 20 such that the resource pool 12 maintains a steady-state
worth.
[0035] The inputs to the controller 18 include a desired
steady-state worth and a stated probability. The stated
probability, which can be steady or variable, represents the
probability that the resource pool 12 will maintain a given value
over a selected interval.
[0036] The stated probability can be used to control the draw
percentage, the evaluation interval and/or the composition of the
resource pool 12. However, in some embodiments, there is no stated
probability input, and both the draw percentage and the extent of
the evaluation interval are predetermined.
[0037] In response to one or more of the foregoing inputs, the
controller 18 causes the resource pool 12 to be drawn upon at a
rate selected such that the value of the resource pool 12 remains
at a desired steady-state value. The controller 18 does so by
controlling the draw rate and/or the composition of the resource
pool 12.
[0038] In some embodiments, the controller 18 applies conventional
actuarial techniques. However, in other embodiments, the controller
18 applies techniques other than actuarial techniques, such as
financial, economic, heuristic investment techniques, and/or
combinations thereof.
[0039] In some cases, one or more of the foregoing inputs can be
set to a default value, in which case it is not necessary to
explicitly provide that input to the controller 18.
[0040] FIG. 2 shows an exemplary procedure carried out by the
controller 18 in maintaining a steady-state worth of the resource
pool 12. The procedure begins with the controller 18 receiving
inputs (step 100). These inputs include the evaluation interval,
the frequency with which draws are to be made, and a desired set
point for the steady-state value. Additional inputs can include an
initial depletion rate, a frequency at which the depletion rate is
to be recalculated, and any constraints on the composition of the
resource pool 12.
[0041] Suitable constraints include those associated with funds in
which asset allocation of the funds is changed over time by the
fund managers in response to changing circumstances. Such funds are
often referred to as "dynamic allocation funds." Dynamic allocation
funds adaptively change allocation on the basis of
circumstances.
[0042] The method described herein can be used in conjunction with
inhomogenous resource pools 12 representative of endowment funds
subject to a dynamic asset allocation process, in which the asset
allocation changes over time in an effort to maintain a
steady-state worth for a given draw rate.
[0043] In another aspect, the method includes determining a first
asset allocation of an endowment fund for a first time period based
on the inputs, and determining a second asset allocation of the
endowment fund, different from the first asset allocation, for a
second time period based on updated inputs.
[0044] The method also includes determining the revised payment
amount made for a predefined period based on the performance of the
endowment. This is advantageous to enable the draw rate and/or
investment strategy for an endowment fund to automatically switch
from higher-risk to lower-risk assets in response to changing
circumstances.
[0045] In some implementations, the endowment fund's assets are
allocated by asset type. Asset types can include domestic and
international equity funds, investment-grade and high-yield fixed
income funds, and short term investments like money market funds,
certificates of deposit (CDs), or treasury bills (T-bills). Other
investment vehicles can include alternative asset types such as
real estate and private investments, or more traditional asset
types such as mutual funds, commingled pools, and/or separate
accounts.
[0046] In some embodiments, the asset type having the largest
percentage of assets over a period of time during existence of the
investment includes domestic equities. In other embodiments, the
endowment includes one or more high risk asset types, such as
international equities and high yield bonds.
[0047] The controller 18 begins by determining whether it is time
to draw from the resource pool 12 (step 100). If it is not yet
time, the controller 18 waits (step 101) and tries again later
(step 100).
[0048] If the controller 18 determines that it is time to draw, it
first retrieves the time series data (step 102). Then, the
controller passes the time series data through a digital filter to
remove high-frequency components representative of short-term
fluctuations in the time-series (step 104). The controller 18 then
determines a draw amount based on the filtered data (step 106).
Typically, this draw amount is a certain percentage of a moving
average of those elements of the time series that are within the
evaluation interval. After determining the draw amount, the
controller 18 outputs data representative of the draw amount to be
used for executing the draw on the resource pool (step 108). This
output data is thus a tangible entity that is relied upon for any
subsequent trades or redemptions that are deemed necessary to
provide the draw amount to an intended beneficiary.
[0049] In those embodiments that implement dynamic reallocation,
the controller 18 determines if there exists a need to change the
composition of the pool 12 by reallocation between its constituent
resources (step 110). If so, the controller outputs data
representative of a recommended reallocation (step 112). This
output data is thus a tangible entity that is relied upon in the
execution of subsequent trades necessary to execute the
reallocation.
[0050] A decision to inspect the resource pool in an effort to
determine whether re-allocation is necessary can be made based on
the passage of time, for example semi-annually, or annually, or on
the extent of fluctuations in, for example, a market index. For
example, one might inspect the resource allocation on a regular
basis. However, if the controller 18 detects a sudden change in a
market index, there may be motivation to inspect the resource pool
to determine whether the sudden change will require a re-allocation
of resources.
[0051] In providing the output data representative of resource
allocation, the controller 18 may apply actuarial, financial,
economic, and/or heuristic methods to actual fluctuations in the
output of the stochastic disturbance source 16 and to forecasts of
future outputs.
[0052] The foregoing description assumes that time series data
exists for the entire evaluation interval. However, this is
generally not the case when the resource pool is first assembled.
In particular, one cannot have thirty-six months of actual time
series data until the resource pool 12 is at least thirty-six
months old.
[0053] To avoid this difficulty, one practice of the method
described herein includes providing synthetic historical data to
fill in the time series at those points in which no real data is
available. Such synthetic data can be obtained on the basis of
historical data representative of market indicators, or historical
data representing the values of the individual resources before
those resources were assembled together into a resource pool.
[0054] In some cases, there exist externalities that limit or
otherwise affect the draw rate. For example, in the case of a
resource pool 12 containing financial instruments, there may be tax
laws that impose constraints on either the maximum amount that may
be drawn, the draw rate, or both.
[0055] To accommodate such externalities, an alternative
implementation, shown in FIG. 3, follows the determination of draw
amount (step 106) with that of calculating any additional
constraints on that amount (step 121). The resulting draw amount is
then adjusted on the basis of any such constraints (step 122)
before being relied upon for executing a draw from the pool 12
(step 108).
[0056] An investor who provides resources in the resource pool 12
would have complete control over those resources. For example, in
some embodiments, the investor can draw resources at any time. In
other embodiments, the investor can add resources to the resource
pool 12. Such changes can easily be accommodated because, in
principle, they are no different from a stochastic disturbance from
the stochastic disturbance source 16.
[0057] The composition of the resource pool 12 is typically a
diversified mix of investments, including mutual funds that invest
in US and foreign stocks, bonds, and cash investments. A realistic
minimum initial value, or initial investment, would be US$100,000,
with a mutual fund expense ratio of 1% and no additional fees.
[0058] It is to be understood that while the invention has been
described in conjunction with the detailed description thereof, the
foregoing description is intended to illustrate and not limit the
scope of the invention, which is defined by the scope of the
appended claims.
[0059] Having described the invention, and a preferred embodiment
thereof, what we claim as new and secured by letters patent is:
* * * * *