Hi everyone,

I am conducting a mediation analysis using the ml_mediation command on Stata which by default is set to REML. I have the following set up:

ml_mediation, dv( AR ) mv( RD ) iv( CP ) l2id( ID ) cv( LaggedAR, Size, Profitability, Growth)

I have already conducted Wooldridge's test for autocorrelation on the 3 equations in my set up for mediation analysis:
MV= IV+ CV1+ CV2+CV3+CV4
DV=IV+ CV2+CV3+CV4
DV=IV+ MV+ CV2+CV3+CV4

Where DV= Dependent Variable
IV= Independent Variable
MV= Mediator Variable
CV= Control variable
Do I need to check for heteroscedasticity? If yes then how can I do so in Stata and will I do it for all 3 equations in my model?