DOMINANCE & RELATIVE WEIGHT ANALYSIS I
DOMINANCE & RELATIVE WEIGHT ANALYSIS I
clear
matrix input Mean = (0,0,0,0,0)
matrix input SD = (1,1,1,1,1)
matrix input Corr = (1, .30, .30, .25, .25 \ .30, 1, .60, .60, .60 \ .30, .60, 1, .60, .60 \ .25, .60, .60, 1, .60 \ .25, .60, .60, .60, 1)
* Now use corr2data to create a pseudo-simulation of the data
corr2data var_y var_x1 var_x2 var_x3 var_x4, seed(0) n(10609) means(Mean) corr(Corr) sds(SD)
* Confirm that all is well
corr var_y var_x1 var_x2 var_x3 var_x4, means
*1.Partial Correlation
regress var_y var_x1 var_x2 var_x3 var_x4
estat esize
pcorr var_y var_x1 var_x2 var_x3 var_x4
*2.Dominance anlaysis
domin var_y var_x1 var_x2 var_x3 var_x4, reg(regress)
*3.Relative weight analysis
domin var_y var_x1 var_x2 var_x3 var_x4, epsilon reg(regress)
*4 Produce standard errors using bootstrapping
bootstrap, reps(1000): domin var_y var_x1 var_x2 var_x3 var_x4, epsilon
estat bootstrap
/*
Reference
Tonidandel, S., & LeBreton, J. M. (2011). Relative Importance Analysis: A Useful Supplement to Regression Analysis. Journal of Business and Psychology, 26(1), 1–9. https://doi.org/10.1007/s10869-010-9204-3
*/
Get in touch.
If you have any additional questions or require further clarification, please, do not hesitate to contact me.
SunjinPak@Gmail.com