Footnote:
Nach Informationen von SSRN wurde die ursprüngliche Fassung des Dokuments February 17, 2023 erstellt
Description:
The literature on p-hacking primarily highlights its prevalence and the need to pay attention to the practical significance of each regressor. A new R package, 'practicalSigni,' reports 13 indexes for p regressors called m1 to m13. The following methods are missing from the literature. (m6) Generalized partial correlation coefficients (GPCC). (m7) extending Psychologists' "Effect Sizes'' to two or more regressors. (m8, m9) Two kernel regression partial derivatives from np and NNS packages. (m10) The NNS boost function of NNS. (m11) re-imagines the regression as a cooperative game (Shapley Value) of forecasting. Two random forest feature importance measures (m12, m13) use out-of-bin (OOB) calculations. Since regression is important for all quantitative sciences, my package offers simple command like report Rank (y, X), that summarize the ranks by newer nonparametric methods (m6 to m13)