Patent Issued for Planning engine for a financial planning system (USPTO 11748814): Empower Annuity Insurance Company of America

2023 SEP 22 (NewsRx) -- By a News Reporter-Staff News Editor at Insurance Daily News -- Empower Annuity Insurance Company of America (Greenwood Village, Colorado, United States) has been issued patent number 11748814, according to news reporting originating out of Alexandria, Virginia, by NewsRx editors.

The patent’s inventors are Cosmano, Brian (Greenwood Village, CO, US), Hersch, Jeremy (Greenwood Village, CO, US), Hough, Sean (Thornton, CO, US), O’Connell, Paul (Wilmington, MA, US), Probetts, Daryl (Highlands Ranch, CO, US).

This patent was filed on April 30, 2020 and was published online on September 5, 2023.

From the background information supplied by the inventors, news correspondents obtained the following quote: “The disclosure relates generally to account management, and more specifically to, a user interface for manipulating account projections based on financial data captured and aggregated from multiple sources.

“Increasingly, individuals are responsible for managing their own personal retirement accounts, which may be supplemented by employer contributions. Individual accountholders may be ill-equipped to optimize retirement accounts, as the lengthy term of the account increases sensitivity to asset allocation, contribution strategies, withdraw strategies, and changes in supplemental retirement benefits. For example, market conditions and/or a projected retirement date may require specific adjustments to the asset allocation of the account. Additionally, contribution and withdrawal policies may change. For example, taxes may be adjusted or assessed differently from year to year. As a result, many participants in financial plans, such as employer-provided 401(k) plans, would benefit from enrollment in a financial planning system that provides enhanced service, such as improved recommendations and visualization tools. However, at least some known financial planning systems, in attempting to provide a more sophisticated set of tools for the user, present a dramatically different and/or more complex user interface as compared to the basic 401(k) plan management interface to which many users are accustomed. As a result, many ordinary participants may be dissuaded from enrolling in, or continuing to stay enrolled in, such enhanced services.

“Moreover, at least some known conventional on-line financial planning systems generate a large number of recommendations that overwhelm the ordinary user, and/or recommendations that are too complex for the ordinary user to grasp, and/or recommendations that result in changes that appear extreme to the ordinary user. As a result, financial planning participants using such known systems may be unable or unwilling to take steps to improve their income in retirement.”

Supplementing the background information on this patent, NewsRx reporters also obtained the inventors’ summary information for this patent: “The following detailed description illustrates embodiments of the disclosure by way of example and not by way of limitation. The description enables one skilled in the art to make and use the disclosure. It also describes several embodiments, adaptations, variations, alternatives, and uses of the disclosure, including what is presently believed to be the best mode of carrying out the disclosure.

“The systems and methods of the disclosure facilitate generating account projections based on financial data captured and aggregated from multiple sources, and alerting accountholders in response to changes in the account projection.

“A GUI computer system for providing a graphical user interface (GUI) is provided. The GUI computer system is in communication with at least one planning engine. The planning engine includes multiple optimization modules for planning significant personal economic events (e.g., investment, savings, retirement). The GUI computer system is configured to capture and centrally store user profile data, including the financial data used by the optimization modules, such as the composition of investment accounts (e.g., employer-provided 401(k) plans). The planning engine is configured to retrieve additional financial data from multiple external data sources, such as asset return projections. Additionally, the GUI computer system may be configured to provide a web interface to capture financial data from the user. For example, the GUI computer system may provide a HTTP API or an interactive web application through which the user can enter data. Additionally, the GUI computer system may be configured to cause the GUI to be displayed by an application installed on a mobile device of a user.

“The planning engine is configured to generate account projections in response to captured financial data. In some embodiments, the planning engine generates account projections (e.g., in near-real-time) based on updated financial data retrieved from the external data sources (e.g., third-party banking institutions, investment institutions) and/or from the user interface. For example, a deferral optimization module generates an updated account projection based on changes in employer contribution formulas, maximum contribution formulas, effective tax rates, current contribution rates, account allocation, project return data, and the like.

“The planning engine is further configured to, during the generation of account projections, generate a portfolio data object for each of a plurality of future years. The portfolio data object is configured to calculate an expected portfolio return across a plurality of asset classes, using an expected asset class return weighted by an assigned asset class weight, and a portfolio standard deviation across the plurality of asset classes, using an asset class standard deviation and an asset class covariance weighted by the assigned asset class weights. After generating the portfolio data object, the planning engine is configured to pass the portfolio data object to a monte carlo return object. The monte carlo return object executes a number of simulations to project a return on the account data over the plurality of future years and outputs a matrix of projected returns over the plurality of future years. While traditional planning engines require the monte carlo algorithm to operate separately on the expected return and standard deviation for each asset class in the portfolio, the return calculation module of the planning engine described herein requires the monte carlo return algorithm to operate only once on the portfolio for each year because the expected portfolio return and portfolio standard deviation are pre-derived from the asset class values by the portfolio data object. Thus, the number of randomized simulations is reduced to one per year, greatly reducing the computational resource intensity required while performing the monte carlo simulations.”

The claims supplied by the inventors are:

“1. A planning engine for a financial planning system, the planning engine comprising at least one processor programmed to: receive, from a graphical user interface (GUI) executed on a user computing device and through a first application programming interface (API), user profile data and account data for a user, wherein the GUI is configured to communicate with a second planning engine via a second API different from the first API; assign, for each of a plurality of future years, an asset class weight to each of a plurality of asset classes associated with the account data; retrieve, for each of the plurality of asset classes, an expected asset class return, an asset class standard deviation, and an asset class covariance; generate a portfolio data object for each of the plurality of future years, the portfolio data object including the assigned asset class weight for each of the plurality of asset classes for the respective year, wherein the portfolio data object is configured to calculate (i) an expected portfolio return across the plurality of asset classes using the expected asset class return weighted by the assigned asset class weight, and (ii) a portfolio standard deviation across the plurality of asset classes using the asset class standard deviation and the asset class covariance each weighted by the assigned asset class weight; pass the portfolio data object for each of the plurality of future years to a monte carlo return object, wherein the monte carlo return object is configured to execute a number of simulations on each portfolio data object using the expected portfolio return and the portfolio standard deviation to project a return on the account data over the plurality of future years; receive, from the monte carlo return object, a matrix having a first dimension equal to a number of the plurality of years and a second dimension equal to the number of simulations, wherein each value in the matrix is the projected return for a corresponding one of the years and a corresponding one of the simulations; and return, to the user computing device, an account projection derived from the matrix, wherein the GUI further includes an interactive control configured to be manipulated by the user to provide modified values for the user profile, and the GUI is further configured to transmit the modified values through an API to update the user profile.

“2. The planning engine according to claim 1, wherein the at least one processor is further programmed to: generate a glidepath data object comprising the portfolio data object for each of the plurality of future years; and pass the portfolio data object for each of the plurality of future years to the monte carlo return object by passing the glidepath data object.

“3. The planning engine according to claim 1, wherein the at least one processor is further programmed to: generate the portfolio data object further configured to verify, prior to passing the portfolio data object to the monte carlo return object, that none of the assigned asset class weights in the portfolio data object is less than zero, and that the sum of the assigned asset class weights in the portfolio data object is one-hundred percent.

“4. The planning engine according to claim 1, wherein the at least one processor is further programmed to: select, from a plurality of sets of initialization parameters, a first set of initialization parameters based on the received account data; and validate the plurality of asset classes associated with the account data against account type data included in the first set of initialization parameters.

“5. The planning engine according to claim 4, wherein the at least one processor is further programmed to: retrieve, from the first set of initialization parameters, an earnings growth rate based on the user profile data; and before returning the account projection, adjust the values in the matrix based on the earnings growth rate.

“6. The planning engine according to claim 4, wherein the at least one processor is further programmed to: receive, from a second user computing device, user profile data and account data for a second user; select, from the plurality of sets of initialization parameters, a second set of initialization parameters based on the account data for the second user; and apply the second set of initialization parameters to the user profile data and the account data for the second user to generate an account projection for the second user.

“7. A planning engine for a financial planning system, the planning engine comprising at least one central processing unit (CPU) and a graphics processing unit (GPU), the planning engine configured to: receive, at the at least one CPU from a graphical user interface (GUI) executed on a user computing device and through a first application programming interface (API), user profile data and account data for a user, wherein the GUI is configured to communicate with a second planning engine via a second API different from the first API; assign, by the at least one CPU for each of a plurality of future years, an asset class weight to each of a plurality of asset classes associated with the account data; generate, by the at least one CPU, a portfolio data object for each of the plurality of future years, the portfolio data object including the assigned asset class weight for each of the plurality of asset classes for the respective year, wherein the portfolio data object is configured to calculate an expected portfolio return across the plurality of asset classes and a portfolio standard deviation across the plurality of asset classes; execute, by the GPU, a number of simulations on the portfolio data object for each of the plurality of future years, using the expected portfolio return and the portfolio standard deviation, to project a return on the account data over the plurality of future years; receive, at the at least one CPU from the GPU, a matrix having a first dimension equal to a number of the plurality of years and a second dimension equal to the number of simulations, wherein each value in the matrix is the projected return calculated by the GPU for a corresponding one of the years and a corresponding one of the simulations; and return, by the at least one CPU to the user computing device, an account projection derived from the matrix, wherein the GUI further includes an interactive control configured to be manipulated by the user to provide modified values for the user profile, and the GUI is further configured to transmit the modified values through an API to update the user profile.

“8. The planning engine according to claim 7, wherein the planning engine is further configured to: generate, by the CPU, a glidepath data object comprising the portfolio data object for each of the plurality of future years; and pass, from the CPU, the portfolio data object for each of the plurality of future years to the GPU by passing the glidepath data object.

“9. The planning engine according to claim 7, wherein the planning engine is further programmed to: generate, by the CPU, the portfolio data object further configured to verify, prior to passing the portfolio data object to the GPU, that none of the assigned asset class weights in the portfolio data object is less than zero, and that the sum of the assigned asset class weights in the portfolio data object is one-hundred percent.

“10. The planning engine according to claim 7, wherein the planning engine is further programmed to: select, by the CPU, from a plurality of sets of initialization parameters, a first set of initialization parameters based on the received account data; and validate, by the CPU, the plurality of asset classes associated with the account data against account type data included in the first set of initialization parameters.

“11. The planning engine according to claim 10, wherein the planning engine is further programmed to: retrieve, by the CPU, from the first set of initialization parameters, an earnings growth rate based on the user profile data; and before returning the account projection, adjust, by the CPU, the values in the matrix based on the earnings growth rate.

“12. The planning engine according to claim 10, wherein the planning engine is further programmed to: receive, from a second user computing device, user profile data and account data for a second user; select, by the CPU, from the plurality of sets of initialization parameters, a second set of initialization parameters based on the account data for the second user; and apply, by the CPU, the second set of initialization parameters to the user profile data and the account data for the second user to generate an account projection for the second user.

“13. The planning engine according to claim 7, wherein the planning engine is further configured to retrieve, by the CPU, an expected asset class return, asset class standard deviation, and asset class covariance, and pass from the CPU, the portfolio data object to the GPU.”

There are additional claims. Please visit full patent to read further.

For the URL and additional information on this patent, see: Cosmano, Brian. Planning engine for a financial planning system. U.S. Patent Number 11748814, filed April 30, 2020, and published online on September 5, 2023. Patent URL (for desktop use only): https://ppubs.uspto.gov/pubwebapp/external.html?q=(11748814)&db=USPAT&type=ids

(Our reports deliver fact-based news of research and discoveries from around the world.)