Integrated risk management process

• US 9,020,857 B2
• Filed: 02/11/2009
• Issued: 04/28/2015
• Est. Priority Date: 02/11/2009
• Status: Active Grant

• Current Assignee: Johnathan C. Mun
• Original Assignee: Johnathan C. Mun
• Inventors: Mun, Johnathan C.
• Primary Examiner(s): Nigh, James D
• Application Number:
  US12/378,169
• Publication Number:
  US 20100205042A1
• Time in Patent Office: 2,267 Days
• Field of Search: 705/7.28, 705/7.11, 705/7.31, 705/59, 705/348
• US Class Current: 705/59
• CPC Class Codes: G06F 17/18 for evaluating statistical data , e.g. average values, frequency distributions, probability functions, regression analysis forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04
• CPC Class Codes: G06F 30/20 Design optimisation, verification or simulation optimisation, verification or simulation of circuit designs G06F30/30
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/0633 Workflow analysis
• CPC Class Codes: G06Q 2220/18 Licensing
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities

Abstract

A method and system allowing the analysis of risk through the use of Monte Carlo simulation, statistical and data analysis, stochastic forecasting, and optimization. The present invention includes novel methods such as the detailed reporting capabilities coupled with advanced analytical techniques, an integrated risk management process and procedures, adaptive licensing technology, and model profiling and storage procedures.

Claims

1. A non-transitory computer-readable medium that stores computer-executable instructions that are executable by a computer processor, the instructions when executed embodying a method that comprises:

using a computer processor in a computing device comprising an operating system, hard drive, motherboard, and Ethernet card to store, in a non-transitory computer-readable medium, a plurality of forecasting models and methods to automatically analyze business risk, related to a company, through an intelligent set of statistical and analytical tests;

using an automatic autoregressive integrated moving average (ARIMA) model to rank said forecasting models from best to worst based on user provided data, so that a user can make an informed decision as to which model to use for a particular set of data, wherein said ranking of models comprises the steps of;

testing said forecasting models for heteroskedastic data by applying White'"'"'s test;

altering any said heteroskedastic data into homoskedastic data by applying a Weighted Least Squares (WLS) approach;

testing said forecasting models for multicollinearity by creating a correlation matrix between the independent variables of said user-provided data, wherein a cross correlation, in said correlation matrix, of greater than 0.75 indicates undesirable multicollinearity, and correcting any said undesirable multicollinearity;

testing said forecasting models for autocorrelation using a Durbin-Watson statistic;

fixing any found autocorrelation in said forecasting models by adding time lags of said found autocorrelation into a regression function and testing for the significance of said time lags;

testing said forecasting models for model misspecification by using White'"'"'s test and a Durbin-Watson statistic;

determining a best fit model for said user-provided data by testing various combinations of p, d, q integers in said ARIMA model in an automated and systematic fashion and applying a plurality of goodness-of-fit statistics to said ARIMA model, wherein said plurality of goodness-of-fit statistics consists of a t-statistic, an F-statistic, an R-squared statistic, and an adjusted R-squared statistic;

representing graphically one or more independent variables contained in said best fit model against one or more dependent variables contained in said best fit model;

using Akaike information criterion on one or more of said various combinations of p, d, q integers in said ARIMA model to determine a Akaike information criterion value;

using Schwarz criterion on one or more of said various combinations of p, d, q integers in said ARIMA model to determine a Schwarz criterion value;

selecting, as the best fit model, the ARIMA model with the lowest Akaike information criterion and said Schwarz criterion values;

calculating a set of autocorrelation statistics from said best fit model;

calculating a set of partial autocorrelation statistics from said best fit model;

creating a result from at least said best fit model, said set of autocorrelation statistics and said set of partial autocorrelation statistics; and

utilizing said results to determine a ranking of said forecasting models.

2. A computer implemented method for automatically analyzing business risk through an intelligent set of statistical and analytical tests, said method comprising:

using a computer processor in a computing device comprising an operating system, hard drive, motherboard, and Ethernet card to store, in a non-transitory computer-readable medium, a plurality of methods to automatically analyze business risk, related to a company, through an intelligent set of statistical and analytical tests;

using an automatic autoregressive integrated moving average (ARIMA) model to rank statistical forecasting models from best to worst, so that a user can make an informed decision as to which model to use, wherein said ranking of models comprises the steps of;

testing said forecasting models for heteroskedastic data by applying White'"'"'s test;

altering any said heteroskedastic data into homoskedastic data by applying a Weighted Least Squares (WLS) approach;

testing said forecasting models for multicollinearity by creating a correlation matrix between the independent variables of said user-provided data, wherein a cross correlation, in said correlation matrix, of greater than 0.75 indicates undesirable multicollinearity, and correcting any said undesirable multicollinearity;

testing said forecasting models for autocorrelation using a Durbin-Watson statistic;

fixing any found autocorrelation in said forecasting models by adding time lags of said found autocorrelation into a regression function and testing for the significance of said time lags;

testing said forecasting models for model misspecification by using White'"'"'s test and using a Durbin-Watson statistic;

determining a best fit model for said user-provided data by testing various combinations of p, d, q integers in said ARIMA model in an automated and systematic fashion and applying a plurality of goodness-of-fit statistics to said ARIMA, wherein said plurality of goodness-of-fit statistics consists of a t-statistic, an F-statistic, an R-squared statistic, and an adjusted R-squared statistic;

representing graphically one or more independent variables contained in said best fit model against one or more dependent variables contained in said best fit model;

using Akaike information criterion on one or more of said various combinations of p, d, q integers in said ARIMA model to determine a Akaike information criterion value;

using Schwarz criterion on one or more of said various combinations of p, d, q integers in said ARIMA model to determine a Schwarz criterion value;

selecting, as the best fit model, the ARIMA model with the lowest Akaike information criterion and said Schwarz criterion values;

calculating a set of autocorrelation statistics from said best fit model;

calculating a set of partial autocorrelation statistics from said best fit model;

creating a result from at least said best fit model, said set of autocorrelation statistics and said set of partial autocorrelation statistics; and

utilizing said results to determine a ranking of said forecasting models.