Integrated Risk Management System

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.

I claim: 
     
         1 . A system for analyzing business risk comprising:
 a computing device comprising a processor communicatively connected to a storage medium, a motherboard, and an Ethernet card;   an operating system stored in a memory of said device configured to provide instructions to said processor;   computer readable instructions residing in a memory of said device, wherein said computer readable instructions comprise a risk simulation module comprising:   
       a plurality of forecasting and risk simulation models and methods, and a plurality of basic econometric models; 
       a basic econometrics module configured to run one or more basic econometric models by (1) identifying input variables from said user provided data and designating at least one independent variable and at least one dependent variable from among said input variables, (2) calculating at least one of the following metrics: R-Squared, Adjusted R-Squared, Multiple R, Standard Error of the Estimates, ANOVA F Statistic, and ANOVA p-Value; (3) testing for regression errors including at least one of: heteroskedasticity, multicollinearity, micronumerosity, lags, leads, and autocorrelation, and (4) adjusting the data to fix any identified regression errors; 
       an autoregressive integrated moving average (ARIMA) module configured to analyze and rank said forecasting and risk simulation models and methods from best to worst based on said adjusted user provided data, by testing various combinations of p, d, and q integers to determine the best-fitting model for the user provided data, wherein one or more of said ranked forecasting and risk simulation models may be selected for use in a simulation; 
       a simulation selection module configured to allow a user to select an active simulation defined by one or more forecasting and/or risk simulation models being applied to a set of input variables derived from the user provided data; 
       a stochastic process forecasting module configured to forecast future values for at least one of equities, assets, interest rates, inflation rates, and commodities using at least one of Brownian motion random walk, mean-reversion, and jump-diffusion; 
       a distribution analysis module configured to generate the probability density function (PDF), cumulative distribution function (CDF), and the inverse cumulative distribution function (ICDF) of distributions calculated in the Risk Simulator; and 
       a Statistical Analyses module comprising a Descriptive Statistics sub-module which includes descriptive statistics functions, a Distributional Fitting sub-module which includes distributional fitting functions, a Histogram and Charts sub-module which includes histogram and chart generating functions, a Hypothesis Testing sub-module which includes hypothesis testing functions to determine the probability that a given hypothesis is true, a Nonlinear Extrapolation sub-module which includes extrapolation functions that extrapolates or extends non-linear data into the future, a Normality Test sub-module which includes functions for determining whether the user provided data set is well-modeled by a normal distribution and how likely it is for a variable underlying the data set to be normally distributed, a Stochastic Process Parameter Estimation sub-module which includes functions for estimating parameters to achieve a best fit regarding characteristics of the user provided data set, a Time-series Autocorrelation sub-module which includes functions for identifying auto correlation as a function of time, and a Trend Line Projection sub-module which includes trend line projection functions. 
     
     
         2 . The system of  claim 1 , wherein the best-fitting model for the user-provided data is determined by applying several goodness-of-fit statistics comprising a t-statistic, F-statistic, R-squared statistic, adjusted R-squared statistic, Durbin-Watson statistic, Akaike Criterion, Schwartz Criterion, and their respective probabilities. 
     
     
         3 . The system of  claim 1 , wherein said testing for regression errors is accomplished by graphically representing each independent variable against the at least one dependent variable. 
     
     
         4 . The system of  claim 1 , wherein said testing for regression errors is accomplished by applying White's test. 
     
     
         5 . The system of  claim 1 , wherein when heteroskedasticity is detected by said testing, said heteroskedasticity is made homoskedastic by using a weighted least squares (WLS) approach. 
     
     
         6 . The system of  claim 1 , wherein Park's test is used to test for and fix heteroskedasticity. 
     
     
         7 . The system of  claim 1 , wherein the test for multicollinearity comprises gauging whether the R-squared value too high and the t-statistics are too low. 
     
     
         8 . The system of  claim 1 , wherein the test for multicollinearity comprises a correlation matrix between the independent variables, wherein when correlation values in said matrix are high the system identifies a potential for multicollinearity. 
     
     
         9 . The system of  claim 8 , wherein the system determines that multicollinearity is severe when the cross correlation between the independent variables has an absolute value that is greater than 0.75. 
     
     
         10 . The system of  claim 1 , wherein the test for multicollinearity comprises the use of a variance inflation factor (VIF). 
     
     
         11 . The system of  claim 1 , wherein the test for autocorrelation comprises graphing the time series of a regression equation's residuals. 
     
     
         12 . The system of  claim 11 , wherein the system determines that autocorrelation exists when said residuals exhibit some cyclicality. 
     
     
         13 . The system of  claim 1 , wherein the test for autocorrelation comprises using the Durbin-Watson statistic. 
     
     
         14 . The system of  claim 13 , wherein the Durbin-Watson statistic is used to identify model misspecification. 
     
     
         15 . The system of  claim 1 , wherein the test for autocorrelation comprises the Breusch-Godfrey test. 
     
     
         16 . The system of  claim 1 , wherein any autocorrelation is fixed by taking the lags of the at least one dependent variable for a relevant period, adding them into the regression function, and testing for their significance. 
     
     
         17 . The system of  claim 1 , wherein the ARIMA module uses an Akaike Information Criterion (AIC) and Schwartz Criterion (SC) to analyze and rank the forecasting and risk simulation models. 
     
     
         18 . The system of  claim 17 , wherein said ARIMA module generates a report comprising autocorrelation (AC) and partial autocorrelation statistics (PAC), wherein said AC, PAC, SC, and AIC are used to help identify the best model.