I have a question regarding calculating the simple effects of a linear regression model with interaction terms. Using STATA I apply the following command (regress FHNIT D1_EU##c.C_Inflat) to which I get results showing that the interaction term between the two independent variables in significant.
Being that the interaction term is significant, I want to look at the simple slopes in order to analyze the interaction of the two variables at specific values. I have tried playing with the commands, however, I cannot seem to get STATA to produce the simple effects. Perhaps this function is not supported. Please let me know how I may be able to produce the simple effects through STATA (or even SPSS).
Hi Eltion,
I have written an ado that generates plots for interaction terms of multiplicative regressions. Just type in ssc install interactplot and look up the help file via help interactplot. I provided some simple and some more advanced examples in the help file, which should help you getting used to the command.
Technical Report interactplot Generate plots for interaction terms of multipl...
Alternatively, you can create marginal effect plots in Stata by using the margins and marginsplot command.
UPDATE:
New version 1.1 of interactplot is now available on ssc. Update via ssc install interactplot, replace. The program can handle simple time-series operators (l., L., d., D., L1., L2., etc.), options in the regression command, and interactions that use the ##-operator.
Example using new features:
. webuse grunfeld
. xtpcse D1.invest c.D1.mvalue##c.D1.kstock, corr(ar1)
. interactplot, cme
Hi Eltion,
I have written an ado that generates plots for interaction terms of multiplicative regressions. Just type in ssc install interactplot and look up the help file via help interactplot. I provided some simple and some more advanced examples in the help file, which should help you getting used to the command.
Technical Report interactplot Generate plots for interaction terms of multipl...
Alternatively, you can create marginal effect plots in Stata by using the margins and marginsplot command.
UPDATE:
New version 1.1 of interactplot is now available on ssc. Update via ssc install interactplot, replace. The program can handle simple time-series operators (l., L., d., D., L1., L2., etc.), options in the regression command, and interactions that use the ##-operator.
Example using new features:
. webuse grunfeld
. xtpcse D1.invest c.D1.mvalue##c.D1.kstock, corr(ar1)
. interactplot, cme
Have you looked at the Holmbeck article from 2002? Article Post-hoc Probing of Significant Moderational and Mediational...
It gives easy to follow calculations in syntax, which I'm sure are easy to do in Stata. Otherwise, you can use Hayes's PROCESS module for SPSS. Another option is Jeremy Dawson's website which allows you to calculate simple slopes. http://www.jeremydawson.co.uk/slopes.htm
That's weird, the double gate (##) should do the trick because it is meant to spare you having also to include the two interacted variables separately (which you need to do if you only used the single gate (#) between them). What version of the software do you use? I have 13 and the ## does exactly what you are after.
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis