topic Marginal structural models in Statistical Procedures https://communities.sas.com/t5/Statistical-Procedures/Marginal-structural-models/m-p/844548#M41840 <P>Hi community,</P><P>&nbsp;</P><P>I’m learning MSM model and find a helpful paper published by SAS:&nbsp;<A href="https://support.sas.com/content/dam/SAS/support/en/books/analysis-of-observational-health-care-data-using-sas/61876_excerpt.pdf" target="_blank" rel="nofollow noopener noreferrer">https://support.sas.com/content/dam/SAS/support/en/books/analysis-of-observational-health-care-data-...</A></P><P>In the example, the key time dependent variable is treatment and weights are calculated with treatment as the dependent variable. If I have three time dependent predictors and want to evaluate their effects, should I calculate weights in separate models with each of the three time dependent predictors as the dependent variable and then run final model with specific weight for the time dependent predictor? Does this mean we can't evaluate&nbsp; the three time dependent predictors in one final model?</P><P>&nbsp;</P><P>Thanks.</P> Wed, 16 Nov 2022 03:23:03 GMT JJ24 2022-11-16T03:23:03Z Marginal structural models https://communities.sas.com/t5/Statistical-Procedures/Marginal-structural-models/m-p/844548#M41840 <P>Hi community,</P><P>&nbsp;</P><P>I’m learning MSM model and find a helpful paper published by SAS:&nbsp;<A href="https://support.sas.com/content/dam/SAS/support/en/books/analysis-of-observational-health-care-data-using-sas/61876_excerpt.pdf" target="_blank" rel="nofollow noopener noreferrer">https://support.sas.com/content/dam/SAS/support/en/books/analysis-of-observational-health-care-data-...</A></P><P>In the example, the key time dependent variable is treatment and weights are calculated with treatment as the dependent variable. If I have three time dependent predictors and want to evaluate their effects, should I calculate weights in separate models with each of the three time dependent predictors as the dependent variable and then run final model with specific weight for the time dependent predictor? Does this mean we can't evaluate&nbsp; the three time dependent predictors in one final model?</P><P>&nbsp;</P><P>Thanks.</P> Wed, 16 Nov 2022 03:23:03 GMT https://communities.sas.com/t5/Statistical-Procedures/Marginal-structural-models/m-p/844548#M41840 JJ24 2022-11-16T03:23:03Z