Message/Author 

Sanjoy posted on Tuesday, October 04, 2005  12:36 am



Dear Professor/s ... using MPlus, can we calculate marginal effect of the explanantory variables, given our endogenous variables are categorical ... precisely, this is our model r by R1R3 b By B1 B3 U on r b X1 r on b X2 b on r X3 U is binary and R and B's are 5 point ordinal, X's share some common element ... I am particularly looking for the marginal effects of "r and b" on U thanks and regards 

bmuthen posted on Saturday, October 08, 2005  8:18 pm



This would have to be done via numerical integration. We do have a little Excel program that we have sent to some users and that we will probably eventually integrate with the Mplus graphics part. 

Sanjoy posted on Saturday, October 08, 2005  11:57 pm



Thank you professor ...is there any way to get that Excel program, or for that matter how the relevant neumerical integration could be done in MPlus thanks and regards 

bmuthen posted on Sunday, October 09, 2005  4:07 pm



I just noticed that you have a reciprocal interaction model  let me first check if the Excel program covers that case. 

Sanjoy posted on Sunday, October 09, 2005  9:46 pm



OK Professor ... that would be great, if that particular program covers the reciprocal situation thanks and regards 


Linda and Bengt, This relates to my other question on the XWITH command. Can the marginal effects of a latent variable interaction created with XWITH be determined in Mplus? Would you be able to make the Excel sheet (and possibly the graphics part mentioned above) available to me (and others)? Thank you. 


I suppose what I am looking for is the elasticity. 


The effect of a latent variables interaction is expressed as a variance contribution in equation (19) of the FAQ http://www.statmodel.com/download/LV%20Interaction.pdf We don't have an Excel program for this. 

Back to top 