|
|
Multivariate multiple regression model |
|
Message/Author |
|
|
Hello Dr. Muthens, Can I preform multivariate multiple regression model with M-plus? If it can, where can i find the corresponding codes for it? Thanks, SJL |
|
|
Yes, see the examples in Chapter 3 of the user's guide on the website. |
|
Farah Rayes posted on Monday, August 28, 2017 - 5:13 am
|
|
|
Hello, I am conducting regression analyses using Mplus. I have two different outcome variables. I also have missing data. I am therefore using ML, while making distributional assumptions about the covariates to prevent the loss of data. However, I obtained different results depending on whether I ran the regressions for the two outcome variables simultaneously (i.e., requesting the results all at once using one program) or separately (i.e., requesting the results for each outcome variable alone; running the program twice). As per the following link (https://stats.idre.ucla.edu/mplus/dae/multivariate-regression-analysis/), the individual coefficients and SEs should be the same using either separate OLS regressions or a multivariate multiple regression. Hence, my questions are: - The coefficients and SE differed slightly. Is this due to the missing data? - If so, would you recommend I run the two regressions simultaneously or run each regression separately? (Suppose I don't want to compare the regression coefficients across equations) Thank you! |
|
|
Send your two outputs to Support along with your license number. |
|
Back to top |
|
|