Systems for adaptively adjusting portfolios based on time series volatility

Abstract

This disclosure provides a system for adaptively adjusting an investment portfolio based on time-series volatility. The system includes a server for periodically acquiring and analyzing multiple pieces of information on multiple financial assets, and terminal devices connected to the server via a network. The server includes an artificial intelligence model that applies the weights of each of multiple sub-time periods (jn) within a historical time period (N) to portfolio theory and calculates an optimized portfolio. The weights of each of the multiple sub-time periods (jn) are calculated using the following formula: where σ is the volatility of each of the multiple financial assets, and jn is a positive integer greater than 2. The optimized portfolio obtained by this system can improve the expected rate of return under the same risk.

Claims

1. A system for adaptively adjusting an investment portfolio based on time-series volatility, comprising: A server for periodically acquiring and analyzing multiple pieces of information on multiple financial assets, the server comprising: an artificial intelligence model for applying a weight of each of multiple sub-time periods (jn) in a historical time period (N) to a portfolio theory and calculating an optimized portfolio, wherein the weight of each of the multiple sub-time periods (jn) is calculated using the following formula: where σ is a volatility of each of the multiple financial assets, and jn is a positive integer greater than 2; and a terminal device connected to the server via a network, wherein the optimized portfolio is connected to and pushed to the terminal device via the network.

2. The system for adaptively adjusting a portfolio based on time-series volatility as described in claim 1, wherein the volatility is measured using a standard deviation of the annualized return of each of the plurality of financial assets.

3. A system for adaptively adjusting a portfolio based on time-series volatility as described in claim 2, wherein the annualized rate of return is [(1 + return / principal)^(1 / number of investment years)]-1.

4. A system for adaptively adjusting a portfolio based on time series volatility as described in claim 1, wherein the portfolio theory includes mean-variance analysis theory, modern portfolio theory, mean-left skewed momentum model, capital asset valuation model, arbitrage pricing theory, risk parity model, risk budgeting model, or Black-Litterman model.

5. A system for adaptively adjusting a portfolio based on time-series volatility as described in claim 1, wherein the plurality of financial assets includes stocks, bonds, or a combination thereof.

6. A system for adaptively adjusting a portfolio based on time series volatility as described in claim 1, wherein the historical time interval (N) is from twice the number of the plurality of sub-time intervals (jn) to a multiple of the number of days in the historical time interval (N).

7. A system for adaptively adjusting a portfolio based on time-series volatility as described in claim 1, wherein the historical time period (N) is at least greater than one trading day.

8. A system for adaptively adjusting a portfolio based on time series volatility as described in claim 1, wherein the number of the plurality of sub-time periods (jn) is 2 to the number of days in the historical time period (N).

9. A system for adaptively adjusting a portfolio based on time-series volatility as described in claim 1, wherein the terminal device includes a smartphone, tablet, notebook computer, desktop computer, or a combination thereof.

Images