Autoeconometrics modeling method
First Claim
Patent Images
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 for modeling various combinations of econometric regression analysis that comprises:
- using a computer processor to store, in a non-transitory computer-readable medium, a plurality of customized and flexible financial options and a plurality of methods for autoeconometric modeling;
presenting to a user, via a software module stored on said tangible computer-readable medium, an interface for entering a dataset into a data grid;
receiving input from said user, wherein said input is comprised of an input dataset for insertion to said data grid, a threshold value and a set of one or more user selected options,wherein said one or more user selected options are comprised of one or more of a linear model, a nonlinear model, a time-series lag and an autoregressive lag,wherein said input dataset is comprised of a plurality of user provided variables;
enumerating a plurality of econometric variables from said plurality of user provided variables,wherein said plurality of econometric variables is greater in total number of variables than said plurality of user provided variables;
assigning a fitness rating for said plurality of econometric variables via analysis routines;
associating a p-value to each variable in said plurality of econometric variables;
storing only said p-values in memory;
eliminating one or more variables from said plurality of econometric variables as insignificant, wherein said one or more variables have a p-value greater than said threshold value;
reanalyzing said plurality of econometric variables via econometric analysis routines;
generating one or more final model reports;
ranking said one or more final model reports by an R-Square coefficient; and
displaying said one or more final model reports in order of ranking to said user.
0 Assignments
0 Petitions
Accused Products
Abstract
A method and system allowing the ability to automatically and systematically run thousands and even millions of combinations and permutations of regression, forecasting and econometric trials to determine the best-fitting predictive model.
-
Citations
16 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 for modeling various combinations of econometric regression analysis that comprises:
-
using a computer processor to store, in a non-transitory computer-readable medium, a plurality of customized and flexible financial options and a plurality of methods for autoeconometric modeling; presenting to a user, via a software module stored on said tangible computer-readable medium, an interface for entering a dataset into a data grid; receiving input from said user, wherein said input is comprised of an input dataset for insertion to said data grid, a threshold value and a set of one or more user selected options, wherein said one or more user selected options are comprised of one or more of a linear model, a nonlinear model, a time-series lag and an autoregressive lag, wherein said input dataset is comprised of a plurality of user provided variables; enumerating a plurality of econometric variables from said plurality of user provided variables, wherein said plurality of econometric variables is greater in total number of variables than said plurality of user provided variables; assigning a fitness rating for said plurality of econometric variables via analysis routines; associating a p-value to each variable in said plurality of econometric variables; storing only said p-values in memory; eliminating one or more variables from said plurality of econometric variables as insignificant, wherein said one or more variables have a p-value greater than said threshold value; reanalyzing said plurality of econometric variables via econometric analysis routines; generating one or more final model reports; ranking said one or more final model reports by an R-Square coefficient; and displaying said one or more final model reports in order of ranking to said user. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
9. A computer implemented method for modeling various combinations of variables via econometric regression analysis:
-
using a computer processor to store, in a non-transitory computer-readable medium, a plurality of customized and flexible financial options and a plurality of methods for autoeconometric modeling; presenting to a user, via a software module stored on said tangible computer-readable medium, an interface for entering a dataset into a data grid; receiving input from said user, wherein said input is comprised of an input dataset for insertion to said data grid, a threshold value and a set of one or more user selected options, wherein said one or more user selected options are comprised of one or more of a linear model, a nonlinear model, a time-series lag and an autoregressive lag, wherein said input dataset is comprised of a plurality of user provided variables; enumerating a plurality of econometric variables from said plurality of user provided variables, wherein said plurality of econometric variables is greater in total number of variables than said plurality of user provided variables; assigning a fitness rating for said plurality of econometric variables via econometric analysis; associating a p-value to each variable in said plurality of econometric variables; storing only said p-values in memory; eliminating one or more variables from said plurality of econometric variables as insignificant, wherein said one or more variables have a p-value greater than said threshold value; reanalyzing said plurality of econometric variables via econometric analysis; generating one or more final model reports; ranking said one or more final model reports by an R-Square coefficient; and displaying said one or more final model reports in order of ranking to said user. - View Dependent Claims (10, 11, 12, 13, 14, 15, 16)
-
Specification