# US 2004/0083152 A1 · Method and system to solve dynamic multi-factor models in finance

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2004/0083152 A1
- Document ID: US20040083152A1
- Status: Published application
- Kind: A1
- Application number: 10/431,838
- Filed: 2003-05-07
- Published: 2004-04-29
- Earliest claimed: 2002-05-07
- Assignees: Markov International Processes LLC
- Inventors: Michael Markov; Vadim Mottl; Ilya Muchnik
- Web page: https://patentgrep.com/patent/US20040083152A1

## CPC

- G06Q 40/06: Asset management; Financial planning or analysis
- G06Q 40/00: Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q 40/03: Credit; Loans; Processing thereof
- G06Q 40/04: Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR

## Abstract

Methods and systems for estimating time-varying factor exposures of either an individual financial instrument or a portfolio of such instruments, through the solution of a constrained multi-criteria dynamic optimization problem, providing an estimation error function and one or more transition error functions to be minimized over a period of time. The factor exposures relay the influence of the factors on the return of the instrument or portfolio. The estimation error function provides the estimation error at each time interval between the return of the asset collection and a sum of products of each factor exposure and its respective factor. Each transition error function provides a transition error of each factor exposure between time intervals. In one embodiment, the constraints can include a budget constraint and non-negativity bounds applying to some or all of the factor exposures. In other embodiments, the method and system can be applied to estimating any time-varying weight that is used in a model, to relay the influence of one or more independent variables on a dependent financial or economic variable, through the solution of a constrained multi-criteria dynamic problem, minimizing estimation error and transition error terms. In other embodiments, the solution of a multi-criteria dynamic problem can be used as part of a method and system to determine structural breakpoints for each factor, and also as part of a method and system for determining optimal parameters to weight the transition error functions and selecting the factors included in the model.

## Claims

### Claim 1

A method of estimating time-varying weights for independent variables at each time interval in a period of time, through a dynamic optimization of a model relating an influence of at least one independent variable on a dependent economic variable, the method comprising the steps of:
receiving data related to the dependent economic variable for the plurality of time intervals;
receiving data related to the at least one independent variable for the plurality of time intervals; and
for each of the plurality of time intervals, determining at least one weight minimizing values of at least two objective functions while meeting at least one constraint on possible values for at least one weight,
wherein each of the at least one weights relays the influence of a respective independent variable on the dependent economic variable,
wherein a first objective function represents an estimation error between the dependent economic variable and a predicted dependent economic variable at each time interval, the predicted dependent economic variable being determined as a function of the weight of each independent variable and its respective independent variable, and
wherein at least one second objective function represents a transition error of each weight between time intervals.

### Claim 2 (depends on 1)

The method according to claim 1, further comprising the step of:
for each of the plurality of time intervals, determining a value of an intercept term minimizing values of the at least two objective functions,
wherein the predicted dependent economic variable is determined as a further function of the intercept term, and
wherein one of the at least one second objective functions represents the transition error of each weight between time intervals, and another second objective function represents a transition error the value of the intercept term between time intervals.

### Claim 3 (depends on 2)

The method according to claim 2, wherein the step of determining the value of the intercept term, includes the step of:
formulating the intercept term as an equivalent weight for an independent variable equal to one; and
grouping the equivalent weight and its respective independent variable with the remaining weights and their respective independent variables in the objective functions.

### Claim 4

A method of estimating time-varying weights for independent variables at each time interval in a period of time, through a dynamic optimization of a model relating an influence of at least one independent variable on a dependent financial variable, the method comprising the steps of:
receiving data related to the dependent financial variable for the plurality of time intervals;
receiving data related to the at least one independent variable for the plurality of time intervals; and
for each of the plurality of time intervals, determining at least one weight minimizing values of at least two objective functions while meeting at least one constraint on possible values for at least one weight,
wherein each of the at least one weights relays the influence of a respective independent variable on the dependent financial variable,
wherein a first objective function represents an estimation error between the dependent financial variable and a predicted dependent financial variable at each time interval, the predicted dependent financial variable being determined as a function of the weight of each independent variable and its respective independent variable, and wherein at least one second objective function represents a transition error of each weight between time intervals.

### Claim 5 (depends on 4)

The method according to claim 4, further comprising the step of:
establishing a price of a financial instrument as the dependent financial variable.

### Claim 6 (depends on 4)

The method according to claim 4, further comprising the step of:
establishing a return of a financial instrument as the dependent financial variable.

### Claim 7 (depends on 4)

The method according to claim 4, further comprising the step of:
establishing the dependent financial variable as a function of one of a price and a return of a security.

### Claim 8 (depends on 7)

The method according to claim 7, wherein the establishing step includes the step of:
establishing the dependent financial variable as a function of the price of the security, the function including a logarithm of the price.

### Claim 9 (depends on 4)

The method according to claim 4, further comprising the step of:
establishing a price of a portfolio as the dependent financial variable.

### Claim 10 (depends on 4)

The method according to claim 4, further comprising the step of:
establishing a return of a portfolio as the dependent financial variable.

### Claim 11 (depends on 4)

The method according to claim 4, further comprising the step of:
establishing the dependent financial variable as a function of one of a price and a return of a portfolio.

### Claim 12 (depends on 11)

The method according to claim 11, wherein the establishing step includes the step of:
establishing the dependent financial variable as a function of the price of the portfolio, the function including a logarithm of the price.

### Claim 13 (depends on 4)

The method according to claim 4, further comprising the step of:
establishing one of an economic and a financial index as one of the at least one independent variables.

### Claim 14 (depends on 4)

The method according to claim 4, further comprising the step of:
establishing one of an economic and a financial index as the dependent financial variable.

### Claim 15

A method of estimating time-varying factor exposures at each time interval in a period of time, through a dynamic optimization of a model relating an influence of at least one factor on a return of an asset collection, the method comprising the steps of:
receiving data related to the return of the asset collection for the plurality of time intervals, the asset collection including at least one asset;
receiving data related to the at least one factor for the plurality of time intervals; and
for each of the plurality of time intervals, determining at least one factor exposure minimizing a value of an objective function while meeting at least one constraint on possible values for the at least one factor exposure,
wherein each of the at least one factor exposures relays the influence of a respective factor on the return of the asset collection,
wherein the objective function includes an estimation error term representing an estimation error at each time interval between the return of the asset collection and a sum of products of each of the at least one factor exposure and its respective factor, and wherein the objective function includes at least one transition error term representing a transition error at each time interval after a first time interval for each of the at least one factor exposure between the time interval and a prior time interval.

### Claim 16 (depends on 15)

The method according to claim 15, wherein the determining step includes the step of:
formulating the objective function as a parameter-weighted sum, the parameter-weighted sum being a sum of a norm of the estimation error term and a parameter-weighted norm of each of the at least one transition error term.

### Claim 17 (depends on 16)

The method according to claim 16, wherein the formulating step includes the step of:
formulating the norm of the estimation error term as a function of a deviation of the estimation error term.

### Claim 18 (depends on 16)

The method according to claim 16, wherein the formulating step includes the step of:
formulating the norm of the estimation error term as a function of a squared deviation of the estimation error term.

### Claim 19 (depends on 16)

The method according to claim 16, wherein the formulating step includes the step of:
formulating the parameter-weighted norm of each of the at least one transition error term as a function of a deviation of the at least one transition error term.

### Claim 20 (depends on 16)

The method according to claim 16, wherein the formulating step includes the step of:
formulating the parameter-weighted norm of each of the at least one transition error term as a function of a squared deviation of the at least one transition error term.

### Claim 21 (depends on 15)

The method according to claim 15, wherein the determining step includes the step of:
formulating the objective function as a parameter-weighted sum, the parameter-weighted sum being a sum of a quadratic norm of the estimation error term and a parameter-weighted quadratic norm of each of the at least one transition error term.

### Claim 22 (depends on 21)

The method according to claim 21, wherein the formulating step includes the step of:
determining a penalty coefficient for a transition error term as the parameter weighting the quadratic norm of the transition error term.

### Claim 23 (depends on 21)

The method according to claim 21, further comprising the step of:
establishing the at least one constraint as a non-negativity bound constraint and a budget constraint on the possible values of at least one factor exposure.

### Claim 24 (depends on 23)

The method according to claim 23, wherein the determining step includes the step of:
using an Interior Point Algorithm to determine the at least one factor exposure.

### Claim 25 (depends on 21)

The method according to claim 21, wherein the determining step includes the steps of:
reformulating the objective function without the at least one constraint, the reformulated objective function without the at least one constraint being equivalent to the objective function with the at least one constraint; and
determining the at least one factor exposure by minimizing the value of the reformulated objective function.

### Claim 26 (depends on 25)

The method according to claim 25, wherein the determining step includes the step of:
using one of a Flexible Least Squares or Generalized Flexible Least Squares algorithm to determine the at least one factor exposure.

### Claim 27 (depends on 21)

The method according to claim 21, further comprising the step of:
at each of the plurality of time intervals, determining a structural breakpoint ratio for each factor exposure, the structural breakpoint ratio being a ratio of a minimum of the parameter-weighted sum over all the time intervals to a minimum of a modified parameters weighted sum, the modified parameter-weighted sum being a sum of a quadratic norm of the estimation error and a parameter-weighted quadratic norm of the at least one transition error over all the time intervals, the modified parameter-weighted sum excluding at least part of a transition error term representing a transition error for the factor exposure between the time interval and a prior time interval.

### Claim 28 (depends on 27)

The method according to claim 27, further comprising the step of:
establishing a structural breakpoint for each of the at least one factor, as occurring at the time interval in the period of time in which the structural breakpoint ratio attains a maximum value.

### Claim 29 (depends on 27)

The method according to claim 27, further comprising the step of:
identifying the time intervals providing probable structural breakpoints for each of the at least one factor, as the time intervals with higher structural breakpoint ratios.

### Claim 30 (depends on 27)

The method according to claim 27, further comprising the step of:
establishing a structural breakpoint for each of the at least one factor, as occurring at the time interval in the period of time in which the structural breakpoint ratio attains a maximum value and exceeds a pre-defined value.

### Claim 31 (depends on 27)

The method according to claim 27, further comprising the step of:
identifying the time intervals providing probable structural breakpoints for each of the at least one factor, as the time intervals in which the structural breakpoint ratio for a factor exceeds a pre-defined value.

### Claim 32 (depends on 21)

The method according to claim 21, further comprising the step of:
determining a cross validation statistic of the model.

### Claim 33 (depends on 32)

The method according to claim 32, wherein the step of determining the cross validation statistic, includes the step of:
testing an accuracy of a predicted return on the asset collection at each time interval.

### Claim 34 (depends on 33)

The method according to claim 33, wherein the testing step includes the step of:
calculating a prediction error of the return on the asset collection at each tested time interval as a difference between an actual return on the asset collection and a predicted return on the asset collection at the tested time interval.

### Claim 35 (depends on 34)

The method according to claim 34, wherein the step of calculating the prediction error at each tested time interval, includes the step of:
determining the predicted return on the asset collection at each tested time interval as a function of predicted factor exposures and their respective factors at the tested time interval.

### Claim 36 (depends on 35)

The method according to claim 35, wherein the step of determining the predicted return on the asset collection at each tested time interval, includes the step of:
creating, for each tested time interval, a reduced dataset to determine the predicted return on the asset collection at the tested time interval, the reduced dataset excluding information relating to the actual return on the asset collection at the tested time interval.

### Claim 37 (depends on 36)

The method according to claim 36, wherein the step of determining the predicted return on the asset collection at the tested time interval, includes the step of:
determining, with the reduced dataset for the tested time interval, the predicted factor exposures minimizing the value of the objective function while meeting the at least one constraint at each time interval.

### Claim 38 (depends on 37)

The method according to claim 37, wherein the step of determining the predicted factor exposures, includes the steps of:
(a) determining the predicted factor exposures minimizing the value of the objective function while meeting the at least one constraint at each time interval except for the tested time interval; and
(b) determining the predicted factor exposures at the tested time interval as a function of at least one of the predicted factor exposures determined in step (a).

### Claim 39 (depends on 38)

The method according to claim 38, wherein the step of determining the predicted factor exposures at the tested time interval, includes the step of:
determining the predicted factor exposures at the tested time interval, as an average of the respective predicted factor exposures at a time interval prior to the tested time interval, and at a time interval subsequent to the tested time interval.

### Claim 40 (depends on 38)

The method according to claim 38, wherein the step of determining the predicted factor exposures at the tested time interval, includes the step of:
determining the predicted factor exposures at the tested time interval, as a function of the respective predicted factor exposures at a time interval prior to the tested time interval, and at a time interval subsequent to the tested time interval.

### Claim 41 (depends on 38)

The method according to claim 38, wherein the step of determining the predicted factor exposures at the tested time interval, includes the step of:
determining the predicted factor exposures at the tested time interval, as a weighted-average of the respective predicted factor exposures at a time interval prior to the tested time interval, and at a time interval subsequent to the tested time interval.

### Claim 42 (depends on 37)

The method according to claim 37, wherein the step of determining the predicted factor exposures, includes the steps of:
formulating a modified objective function, the modified objective function including an estimation error term for each time interval except for the tested time interval, and at least one transition error term for each time interval including the tested time interval; and
determining, with the reduced dataset for the tested time interval, the predicted factor exposures minimizing the value of the modified objective function while meeting the at least one constraint at each time interval.

### Claim 43 (depends on 34)

The method according to claim 34, wherein the step of determining the cross validation statistic, includes the step of:
calculating a quadratic norm of the prediction errors of the return on the asset collection over the period of time.

### Claim 44 (depends on 43)

The method according to claim 43, wherein the step of calculating the quadratic norm includes the step of:
summing up over all tested time intervals the square of the difference between the actual return on the asset collection and the predicted return on the asset collection at each time interval, and dividing the sum by a number of the tested time intervals in the period of time.

### Claim 45 (depends on 43)

The method according to claim 43, further comprising the step of:
determining a Predicted R-Squared Statistic as a function of the quadratic norm of the prediction errors and a variance of squares of the returns on the asset collection over the period of time.

### Claim 46 (depends on 34)

The method according to claim 34, wherein the step of determining the cross validation statistic, includes the step of:
calculating an absolute deviation metric of all the prediction errors of the return on the asset collection over the period of time.

### Claim 47 (depends on 46)

The method according to claim 46, wherein the step of calculating the absolute deviation metric includes the step of:
summing up over all tested time intervals the absolute value of the difference between the actual return on the asset collection and the predicted return on the asset collection, and dividing the sum by a number of the tested time intervals in the period of time.

### Claim 48 (depends on 46)

The method according to claim 46, further comprising the step of:
determining a Predicted R Statistic as a function of the cross validation statistic and a variance of the return on the asset collection over the period of time.

### Claim 49 (depends on 32)

The method according to claim 32, further comprising the step of:
determining a value for the parameter weighting the quadratic norm of at least one transition error term in the parameter-weighted sum, as the value minimizing the cross validation statistic for the model.

### Claim 50 (depends on 32)

The method according to claim 32, further comprising the step of:
selecting the at least one factor for the model from possible factors as a function of the cross validation statistic of the model.

### Claim 51 (depends on 50)

The method according to claim 50, wherein the selecting step includes the step of:
selecting a set of at least one factor for the model, as the set minimizing the cross validation statistic for the model.

### Claim 52 (depends on 50)

The method according to claim 50, wherein the selecting step includes the steps of:
selecting at least one factor from possible factors for inclusion in a base model; and
determining a cross validation statistic of the base model.

### Claim 53 (depends on 52)

The method according to claim 52, wherein the step of selecting the at least one factor for the model includes the steps of:
determining a cross validation statistic of each of the possible factors; and
selecting the factors for inclusion in the base model as a function of the cross validation statistic of each possible factor.

### Claim 54 (depends on 52)

The method according to claim 52, wherein the step of selecting the at least one factor for the model includes the steps of:
(a) creating a test model with a composition of factors different from the factors included in the base model;
(b) determining a change in a cross validation statistic from the base model to the test model; and
(c) conforming a composition of the base model to the composition of the test model if an improvement in the cross validation statistic is determined in step (b).

### Claim 55 (depends on 54)

The method according to claim 54, wherein the creating step includes the step of:
adding at least one possible factor to the factors included the base model, in order to form the composition of the test model.

### Claim 56 (depends on 54)

The method according to claim 54, wherein the creating step includes the step of:
removing at least one factor from the factors included in the base model, in order to form the composition of the test model.

### Claim 57 (depends on 54)

The method according to claim 54, wherein the step of selecting the at least one factor includes the step of:
repeating steps (a) through (c) until a lack of improvement is determined in the cross validation statistic.

### Claim 58 (depends on 54)

The method according to claim 54, wherein the step of selecting the at least one factor includes the step of:
repeating steps (a) through (c) until the cross validation statistic of the test model reaches a threshold value.

### Claim 59 (depends on 15)

The method according to claim 15, wherein the determining step includes the step of:
formulating the at least one transition error term as a function determining, at each time interval, a difference between a value of an estimated factor exposure for its respective factor at the time interval, and a value of the factor exposure induced by a change in the respective factor between the time interval and a prior time interval.

### Claim 60 (depends on 59)

The method according to claim 59, wherein the formulating step includes the step of:
scaling the at least one transition error term functions.

### Claim 61 (depends on 60)

The method according to claim 60, wherein the scaling step includes the step of:
incorporating at least one of the at least one constraints into one of the at least one transition error term functions.

### Claim 62 (depends on 61)

The method according to claim 61, wherein the incorporating step includes the step of:
incorporating a budget constraint into the transition error term function, the budget constraint requiring the factor exposures in the model to sum to one.

### Claim 63 (depends on 59)

The method according to claim 59, wherein the determining step includes the step of:
using an iterative algorithm to determine the factor exposures, a solution from a prior iteration of a factor exposure for a prior time interval, being used in a subsequent iteration as the factor exposure for the prior time interval in the transition error term.

### Claim 64 (depends on 63)

The method according to claim 63, wherein the determining step includes the step of:
continuing iterations until the values for the factor exposures converge between consecutive iterations.

### Claim 65 (depends on 15)

The method according to claim 15, further comprising the step of:
defining one of the at least one constraint as a time varying inequality constraint representing prior information about the factor exposures.

### Claim 66 (depends on 65)

The method according to claim 65, wherein the defining step, includes the step of:
defining the time varying bound constraint as restricting possible estimates of the factor exposures to non-negative values.

### Claim 67 (depends on 15)

The method according to claim 15, further comprising the step of:
defining one of the at least one constraint as a general linear constraint relating to a structure of the factor exposures.

### Claim 68 (depends on 67)

The method according to claim 67, wherein the defining step, includes the step of:
defining a budget constraint as the general linear constraint, the budget constraint requiring estimated factor exposures for a subset of factors in the model to sum to one, the subset including at least one of the factors in the model.

### Claim 69 (depends on 67)

The method according to claim 67, wherein the defining step, includes the step of:
defining a budget constraint as the general linear constraint, the budget constraint requiring estimated factor exposures for all the factors in the model to sum to one.

### Claim 70 (depends on 15)

The method according to claim 15, further comprising the step of:
establishing a budget constraint on the possible values of the exposure factors for a subset of the factors in the model, each factor in the subset corresponding to returns of a subset of instruments in a portfolio, the portfolio being the asset collection.

### Claim 71 (depends on 70)

The method according to claim 70, wherein the establishing step includes the step of:
establishing the factors excluded from the subset as the factors corresponding to returns of the hedging instruments in the portfolio.

### Claim 72 (depends on 71)

The method according to claim 71, wherein the step of establishing the excluded factors, includes the step of:
establishing a currency forward rate as an excluded factor.

### Claim 73 (depends on 15)

The method according to claim 15, further comprising the step of:
establishing a hedging constraint on the possible values of the factor exposures for a subset of the factors in the model, each factor in the subset corresponding to returns of a subset of hedging instruments in a portfolio, the portfolio being the asset collection.

### Claim 74 (depends on 73)

The method according to claim 73, wherein the establishing step includes the step of:
formulating the hedging constraint as specifying a hedge ratio within a range of time intervals.

### Claim 75 (depends on 73)

The method according to claim 73, wherein the establishing step includes the step of:
formulating the hedging constraint as including a short-selling constraint, the short-selling constraint controlling a sum of negative factor exposures for the subset.

### Claim 76 (depends on 25)

The method according to claim 25, wherein the establishing step includes the step of:
defining the subset of all of the at least one factor as including a currency index.

### Claim 77 (depends on 15)

The method according to claim 15, further comprising the step of:
selecting one of a security and a portfolio of securities as the asset collection for the model.

### Claim 78 (depends on 77)

The method according to claim 77, wherein the selecting step includes the step of:
identifying at least one economic factor as the at least one factor in the model influencing the return of the one of the security and the portfolio of securities.

### Claim 79 (depends on 78)

The method according to claim 78, wherein the identifying step includes the step of:
identifying one of industrial production, inflation, interest rates, money supply and business cycle indices as one of the at least one economic factor influencing the return of the asset collection.

### Claim 80 (depends on 77)

The method according to claim 77, wherein the selecting step includes the step of:
identifying a function of at least one of industrial production, inflation, interest rates, money supply and business cycle indices as one of the at least one factor influencing the return of the asset collection.

### Claim 81 (depends on 15)

The method according to claim 15, further comprising the step of:
selecting a generic market index as one of the at least one factors in the model influencing the return on the asset collection.

### Claim 82 (depends on 81)

The method according to claim 81, wherein the selecting step includes the step of:
identifying one of a bond index, equity index, economic sector index, country index and currency index as the generic market index.

### Claim 83 (depends on 15)

The method according to claim 15, further comprising the step of:
selecting a function of at least one market index as one of the at least one factor influencing the return of the asset collection.

### Claim 84

A method of estimating time-varying factor exposures at each time interval in a period of time, through a dynamic optimization of a model relating an influence of at least one factor on a return of an asset collection, the method comprising the steps of:
receiving data related to the return of the asset collection for the plurality of time intervals, the asset collection including at least one asset;
receiving data related to the at least one factor for the plurality of time intervals; and
for each of the plurality of time intervals, determining at least one factor exposure minimizing a value of an objective function while meeting at least one constraint on possible values for the at least one factor exposure,
wherein each of the at least one factor exposures relays the influence of a respective factor on the return of the asset collection,
wherein the objective function includes an estimation error term representing an estimation error at each time interval between the return of the asset collection and a sum of products of each of the at least one factor exposure and its respective factor, and
wherein the objective function includes at least one transition error term representing a transition error at each time interval before a last time interval, for each of the at least one factor exposure between the time interval and a subsequent time interval.

### Claim 85

A method of determining structural breakpoints for factors influencing the return of an asset collection over a period of time, the asset collection including at least one asset, and the period of time including a plurality of time intervals, the method comprising the steps of:
receiving data related to the return of the asset collection for the plurality of time intervals, the asset collection including at least one asset;
receiving data related to the at least one factor for the plurality of time intervals;
for each of the plurality of time intervals, determining at least one factor exposure minimizing a value of a parameter-weighted sum, each of the at least one factor exposure relaying the influence of a respective factor on the return of the asset collection, the parameter-weighted sum being a sum of a quadratic norm of an estimation error term and at least one parameter-weighted quadratic norm of a transition error term, the estimation error term representing an estimation error at each time interval between the return of the asset collection and a sum of products of each of the at least one factor exposure and its respective factor, and each transition error term representing transition error at each time interval after a first time interval for each of the at least one factor exposure between the time interval and a prior time interval; and
at each of the plurality of time intervals, determining a structural breakpoint ratio for each factor exposure, the structural breakpoint ratio being a ratio of a minimum of the parameter-weighted sum over all the time intervals to a minimum of a modified parameter-weighted sum, the modified parameter-weighted sum being a sum of a quadratic norm of the estimation error term and a parameter-weighted quadratic norm of the at least one transition error term over all the time intervals, the modified parameter-weighted sum excluding at least part of a transition error term representing a transition error of the factor exposure between the time interval and a prior time interval.

### Claim 86 (depends on 85)

The method according to claim 85, further comprising the step of:
establishing a structural breakpoint for each of the at least one factor, as occurring at the time interval in the period of time in which the structural breakpoint ratio attains a maximum value.

### Claim 87 (depends on 85)

The method according to claim 85, further comprising the step of:
establishing a structural breakpoint for each of the at least one factor, as occurring at the time interval in the period of time in which the structural breakpoint ratio attains a maximum value and exceeds a pre-defined value.

### Claim 88

A method of configuring a model relating factor exposures for each of at least one factor and its respective factor on a return of an asset collection over a period of time, the period of time including a plurality of time intervals, the asset collection including at least one asset, comprising the step of:
receiving data related to an actual return of the asset collection for the plurality of time intervals, each tested time interval being a particular time interval;
receiving data related to the at least one factor for the plurality of time intervals;
creating a reduced dataset for each tested time interval excluding the data related to the actual return of the asset collection at the tested time interval;
determining with each reduced dataset, for each tested time interval, a predicted return of the asset collection as a function of a set of predicted factor exposures and their respective factors, the predicted factor exposures relaying a predicted influence of a respective factor on the actual return of the asset collection; and
determining a cross validation statistic over the period of time as a function of a difference between the actual return of the asset collection and the predicted return of the asset collection at each time interval, wherein the set of the predicted factor exposures is determined with the reduced dataset by determining, at each time interval, the predicted factor exposures minimizing a sum of a quadratic norm of an estimation error term and at least one parameter-weighted transition error term, the estimation error term representing an estimation error at each time interval, except for the tested time interval, between the actual return of the asset collection and a sum of products of each predicted factor exposure and its respective factor, the at least one transition error term representing a transition error at each time interval after a first time interval for each of the at least one factor exposure between the time interval and a prior time interval.

### Claim 89 (depends on 88)

The method according to claim 88, further comprising the steps of:
formulating a model with a set of factors influencing the actual return of the asset collection, and a parameter to weight one of the transition error terms in the sum of the quadratic norm of the estimation error term and the at least one parameter-weighted transition error term, the set of factors including at least one factor.

### Claim 90 (depends on 89)

The method according to claim 89, wherein the formulating step includes the step of:
determining the parameter as a value minimizing the cross validation statistic for the model.

### Claim 91 (depends on 89)

The method according to claim 89, wherein the formulating step includes the step of:
selecting the at least one factor in the set as a function of the cross validation statistic of the model.

### Claim 92

A method for evaluating a performance of an asset collection over a period of time, the asset collection including at least one asset, the period of time including a plurality of time intervals, the method comprising the step of:
receiving information relating an influence of each factor in a set including at least one factor on a return of the asset collection at each time interval,
wherein the influence of each factor is determined as a function of the factor exposures of each respective factor, at least one factor exposure being determined at each time interval minimizes a value of an objective function while meeting at least one constraint on possible values for the at least one factor exposure,
wherein the objective function includes an estimation error term representing an estimation error at each time interval between the return of the asset collection and a sum of products of each of the at least one factor exposure and its respective factor, and
wherein the objective function includes at least one transition error term representing a transition error at each time interval after a first time interval in the period of time, for each factor exposure between the time interval and a prior time interval.

### Claim 93 (depends on 92)

The method according to claim 92, further comprising the step of:
determining: an investment style of a manager of the asset collection as a function of the received information.

### Claim 94

A method for evaluating a performance of an asset collection over a period of time, the asset collection including at least one asset, the period of time including a plurality of time intervals, the method comprising the step of:
providing information relating an investment style of a manager of the asset collection as a function of information relating an influence of each factor in a set including at least one factor on a return of the asset collection at each time interval,
wherein the influence of each factor is determined as a function of the factor exposures of each respective factor, at least one factor exposure being determined at each time interval minimizes a value of an objective function while meeting at least one constraint on possible values for the at least one factor exposure,
wherein the objective function includes an estimation error term representing an estimation error at each time interval between the return of the asset collection and a sum of products of each of the at least one factor exposure and its respective factor, and
wherein the objective function includes at least one transition error term representing a transition error at each time interval after a first time interval in the period of time, for each factor exposure between the time interval and a prior time interval.

### Claim 95

A computer system for estimating time-varying factor exposures at each time interval in a period of time, through a dynamic optimization of a model relating an influence of at least one factor on a return of an asset collection, comprising:
at least one processor, configured to:
receive data related to the return of the asset collection for the plurality of time intervals, the asset collection including at least one asset;
receive data related to the at least one factor for the plurality of time intervals, and
for each of the plurality of time intervals, determine at least one factor exposure minimizing a value of an objective function while meeting at least one constraint on possible values for the at least one factor exposure,
wherein each of the at least one factor exposures relays the influence of a respective factor on the return of the asset collection,
wherein the objective function includes an estimation error term representing an estimation error at each time interval between the return of the asset collection and a sum of products of each of the at least one factor exposure and its respective factor, and
wherein the objective function includes at least one transition error term representing a transition error at each time interval after a first time interval for each of the at least one factor exposure between the time interval and a prior time interval.

### Claim 96 (depends on 95)

The computer system according to claim 95, further comprising:
at least one database module; and
at least one terminal, the processor being operative with the at least one database module and the at least one terminal.

### Claim 97 (depends on 96)

The computer system according to claim 96, wherein the at least one database is configured to provide to the processor the data related to the return of the asset collection and the data related to the at least one factor for the plurality of time intervals.

### Claim 98 (depends on 96)

The computer system according to claim 96, further comprising:
a server housing the processor and the at least one database module; and
a network connecting the at least one database module and the processor with the at least one terminal.

### Claim 99 (depends on 95)

The computer system according to claim 95, wherein the at least one processor includes a first processor and a second processor parallel to the first processor.

### Claim 100 (depends on 99)

The computer system according to claim 99, wherein the first processor operates with the second processor, each processor configured to at least one of receive the data related to the return of the asset collection, receive the data related to the at least one factor and determine the at least one factor exposure for each time interval.

### Claim 101 (depends on 100)

The computer system according to claim 100, wherein the first processor is configured to receive the data related to the return of the asset collection and the data related to the at least one factor, and the second processor is configured to determine the at least one factor exposure for each time interval.

### Claim 102

A computer system for evaluating a performance of an asset collection over a period of time, the asset collection including at least one asset, the period of time including a plurality of time intervals, the computer system comprising at least one processor configured to:
receive information relating an investment style of a manager of the asset collection as a function of an influence of each factor in a set including at least one factor on a return of the asset collection at each time interval,
wherein the influence of each factor is determined as a function of the factor exposures of each respective factor, at least one factor exposure being determined at each time interval minimizes a value of an objective function while meeting at least one constraint on possible values for the at least one factor exposure,
wherein the objective function includes an estimation error term representing an estimation error at each time interval between the return of the asset collection and a sum of products of each of the at least one factor exposure and its respective factor, and
wherein the objective function includes at least one transition error term representing a transition error at each time interval after a first time interval in the period of time, for each factor exposure between the time interval and a prior time interval.

### Claim 103 (depends on 102)

The computer system according to claim 102, wherein the processor is configured to provide the information to an end-user.

### Claim 104 (depends on 102)

The computer system according to claim 102, wherein the processor is configured to receive the information from a server through an internet connection.

### Claim 105 (depends on 102)

The computer system according to claim 102, wherein the processor is configured to receive the information from a server through a private network connection.

### Claim 106 (depends on 102)

The computer system according to claim 102, wherein the processor is configured to receive the information from a web site.

### Claim 107

A computer program product for use with a system evaluating a performance of an asset collection, the computer program product comprising computer usable medium having computer readable program code embodied in the medium for causing a computer to:
receive data related to the return of the asset collection for the plurality of time intervals, the asset collection including at least one asset;
receive data related to the at least one factor for the plurality of time intervals, and
for each of the plurality of time intervals, determine at least one factor exposure minimizing a value of an objective function while meeting at least one constraint on possible values for the at least one factor exposure,
wherein each of the at least one factor exposures relays the influence of a respective factor on the return of the asset collection,
wherein the objective function includes an estimation error term representing an estimation error at each time interval between the return of the asset collection and a sum of products of each of the at least one factor exposure and its respective factor, and wherein the objective function includes at least one transition error term representing a transition error at each time interval after a first time interval for each of the at least one factor exposure between the time interval and a prior time interval.

### Claim 108

An article of manufacture comprising an information storage medium encoded with a computer-readable data structure adapted for use in evaluating over the Internet a performance of an asset collection, said data structure comprising:
at least one data field with information related to a return of an asset collection for a plurality of time intervals in a period of time, the asset collection including at least one asset; at least one data field with information related to at least one factor for the plurality of time intervals, the at least one factor influencing the return of the asset collection; and
at least one data field with information related to at least one factor exposure for the plurality of time intervals, the at least one factor exposure relaying the influence of a respective factor on the return of the asset collection,
wherein the at least one factor exposure at each time interval is determined by minimizing a value of an objective function while meeting at least one constraint on possible values for the at least one factor exposure,
wherein the objective function includes an estimation error term representing an estimation error at each time interval between the return of the asset collection and a sum of products of each of the at least one factor exposure and its respective factor, and
wherein the objective function includes at least one transition error term representing a transition error at each time interval after a first time interval for each of the at least one factor exposure between the time interval and a prior time interval.

### Claim 109

An article of manufacture comprising a propagated signal adapted for use in a method of estimating time-varying factor exposures at each time interval in a period of time, through a dynamic optimization of a model relating an influence of at least one factor on a return of an asset collection,
a. the method comprising the steps of:
receiving data related to the return of the asset collection for the plurality of time intervals, the asset collection including at least one asset;
receiving data related to the at least one factor for the plurality of time intervals; and
for each of the plurality of time intervals, determining at least one factor exposure minimizing a value of an objective function while meeting at least one constraint on possible values for the at least one factor exposure,
wherein each of the at least one factor exposures relays the influence of a respective factor on the return of the asset collection,
wherein the objective function includes an estimation error term representing an estimation error at each time interval between the return of the asset collection and a sum of products of each of the at least one factor exposure and its respective factor, and
wherein the objective function includes at least one transition error term representing a transition error at each time interval after a first time interval for each of the at least one factor exposure between the time interval and a prior time interval; and
b. the signal encoded with machine-readable information relating to the asset collection.

### Claim 110 (depends on 109)

The article of manufacture according to claim 109, wherein the information includes information relating to the return of the asset collection for the plurality of time intervals.

### Claim 111 (depends on 109)

The article of manufacture according to claim 109, wherein the information includes information relating to the at least one factor for the plurality of time intervals.

### Claim 112 (depends on 109)

The article of manufacture according to claim 109, wherein the information includes information relating to the plurality of time intervals.

### Claim 113 (depends on 109)

The article of manufacture according to claim 109, wherein the information includes information relating to the factor exposures for the plurality of time intervals.

### Claim 114 (depends on 109)

The article of manufacture according to claim 109, wherein the information includes information relating to the objective function.

### Claim 115 (depends on 109)

The article of manufacture according to claim 109, wherein the information includes information relating to the estimation error term.

### Claim 116 (depends on 109)

The article of manufacture according to claim 109, wherein the information includes information relating to the at least one transition error term.

### Claim 117

A method of estimating time-varying factor exposures at each time interval (βt(i)) in a period of time (t=1, . . . , N), through a dynamic optimization of a model relating an influence of at least one factor (xt(i), i=1, . . . , n factors) on a return of an asset collection (yt), the method comprising the steps of:
receiving data related to the return of the asset collection (yt) for the plurality of time intervals, the asset collection including at least one asset;
receiving data related to the at least one factor (xt(i), i=1, . . . , n factors) for the plurality of time intervals; and
for each of the plurality of time intervals, determining at least one factor exposure (βt(i)) minimizing a value of an objective function while meeting at least one constraint on possible values for the at least one factor exposure,
wherein each of the at least one factor exposures (βt(i)) relays the influence of a respective factor (xt(i)) on the return of the asset collection (yt),
wherein the objective function and constraints are formulated as the following constrained multi-criteria dynamic optimization problem containing m+2 objective functions to be minimized, associated, respectively, with the estimation error vector and m+1 transition error vectors in certain norms:
44 { ⁢ min ( α 1 , … ⁢ , α N ) ( β 1 , … ⁢ , β N ) ⁢ [LeftDoubleBracketingBar] e [ 1 , … ⁢ , N } [RightDoubleBracketingBar] , min ( α 1 , … ⁢ , α N ) ⁢ [LeftDoubleBracketingBar] w 0 , [ 2 , … ⁢ , N ] [RightDoubleBracketingBar] , min ( β 1 , … ⁢ , β N ) ⁢ [LeftDoubleBracketingBar] w 1 , [ 2 , … ⁢ , N ] [RightDoubleBracketingBar] , … ⁢ , min ( β 1 , … ⁢ , β N ) ⁢ [LeftDoubleBracketingBar] w m , [ 2 , … ⁢ , N ] [RightDoubleBracketingBar] , ⁢ e [ 1 , … ⁢ , N ] = ( e t = y t - α t - x t T ⁢ β t , t = 1 , … ⁢ , N ) ⁢ ( N - dimensional ⁢ ⁢ vector ⁢ ⁢ of ⁢ ⁢ estimation ⁢ ⁢ errors ) , w 0 , [ 2 , … ⁢ , N ] = ( w 0 , t = [LeftBracketingBar] α t - α t - 1 [RightBracketingBar] , t = 1 , … ⁢ , N ) ⁢ w j , [ 2 , … ⁢ , N ] = ( w j , t = [LeftDoubleBracketingBar] β t - β ¨ t [RightDoubleBracketingBar] , t = 1 , … ⁢ , N ) , ⁢ j = 1 , … ⁢ , m ⁢ } ⁢ ( N - dimensional ⁢ ⁢ vector ⁢ ⁢ of ⁢ ⁢ estimation ⁢ ⁢ errors ) , β j , t = f j ⁡ ( x t , … ⁢ , x t - k j ; β t - 1 , … ⁢ , β t - k j ) ⁢ ( transition ⁢ ⁢ equations ) , [LeftDoubleBracketingBar] e [ 1 , … ⁢ , N } [RightDoubleBracketingBar] , [LeftDoubleBracketingBar] w j , [ 2 , … ⁢ , N ] [RightDoubleBracketingBar] ⁢ ⁢ and ⁢ [RightDoubleBracketingBar] ⁢ β t ( i ) - β ¨ t ( i ) [RightDoubleBracketingBar] ⁢ ⁢ are ⁢ ⁢ norms ⁢ ⁢ of ⁢ ⁢ any ⁢ ⁢ kinds , [AutoRightMatch]
subject to constraints:
45 { ⁢ G t ⁢ β _ t + h t ≥ 0 ⁢ ⁢ ( inequality ⁢ ⁢ constraints ) , ⁢ F t ⁢ β _ t + c t = 0 ⁢ ⁢ ( equality ⁢ ⁢ constraints ) , β _ t = ( α t , β t ) . [AutoLeftMatch]
