Hello All stata list members,
How can i do it in stata if my two variables are x1 ,x2 and its Product is z, then how can i calculate the residual centering.
( Under most circumstances, mean centering is an adequate solution to the collinearity problem. At times, however, the mean-centered product or powered term will still have some degree of correlation with its first-order variables that can
influence the partial regression coefficients. To remedy this lack of complete orthogonality with the mean-centering approach, a simple two-step regression technique called residual centering can be used that ensures full orthogonality between a product term and its first-order effects (Lance, 1988). This technique is also generalizable to powered terms.
Residual centering (i.e., orthogonalizing) is a comparable alternative approach to mean centering that also serves to eliminate nonessential multicollinearity in regression analyses. Residual centering, as originally suggested by Lance (1988), is
essentially a two-stage OLS procedure in which a product term is regressed onto its respective first-order effect(s). The residuals of this regression are then used to represent the interaction effect. The variance of this new orthogonalized interaction term contains the unique variance that fully represents the interaction effect, independent of the first-order effect variance (as well as general error or unreliability).
reference: On the Merits of Orthogonalizing Powered and Product Terms: Implications for Modeling Interactions Among Latent Variables
Article in Structural Equation Modeling A Multidisciplinary Journal · December 2006
DOI: 10.1207/s15328007sem1304_1
Looking forward for your kind reply.