CPC G06F 8/77 (2013.01) [G06Q 10/06393 (2013.01); G06Q 40/12 (2013.12)] | 6 Claims |
1. A computer-implemented method of analyzing a computer software codebase, comprising steps performed by one or more computer systems of:
(a) storing custom fitted statistical models in a data store, each custom fitted statistical model calibrated for a different single codebase and created by applying statistical regression techniques to code quality metrics, design quality metrics, and/or test quality metrics independent variables and software economic outcome dependent variables for a codebase, wherein the custom fitted statistical models provide estimates for low-level software economic parameters;
(b) retrieving said custom fitted statistical models from the data store, merging said custom fitted statistical models to generate a standard fitted statistical model for another codebase, and storing the standard fitted statistical model in the data store; and
(c) retrieving said standard fitted statistical model from the data store, based on said standard fitted statistical model, making defect density or developer productivity projections for said another codebase, and outputting the defect density or developer productivity projections.
|