Dear statisticians,
I am currently working in logistic regression model. I would like the output to include marginal effects. Does the spss have the option for marginal effects?
I am using SPSS 24.
There are multiple ways to calculate marginal effects, so you'd have to specify which you want. Some are simple; for example, calculating the marginal effect at the mean (hold all independent variables to their mean value, then calculate the associated increase in probability for a one-unit increase in the IV of interest) are easy and don't need special functions.
If you're willing to use R, some solutions have already been implemented, see e.g. the bottom half of http://www.mypolyuweb.hk/~sjpolit/logisticregression.html
If you estimate your model via GENLIN (with logit link and binomial error distribution), you'll have access to an EMMEANS (estimated marginal means) sub-command that may give you what you need.
If you also want contrasts, it might be more convenient to estimate your model via NOMREG, which has a TEST sub-command.
HTH.
http://www.ibm.com/support/knowledgecenter/SSLVMB_22.0.0/com.ibm.spss.statistics.reference/spss/regression/syn_nomreg_test.htm
http://www.ibm.com/support/knowledgecenter/SSLVMB_22.0.0/com.ibm.spss.statistics.reference/spss/advanced/syn_genlin_emmeans.htm
There ar a number of various marginal effects; Stata calculates all of them :-)
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis