2
$\begingroup$

I ran a probit regression using the following code:

    m1<-glmer(Success~Name.Origin+(1|Job.ID),family=binomial(link="probit"))

However, I am now unsure how to compute the marginal effects and their corresponding standard errors?

$\endgroup$

closed as off-topic by mdewey, whuber Dec 24 '16 at 14:56

This question appears to be off-topic. The users who voted to close gave this specific reason:

  • "This question appears to be off-topic because EITHER it is not about statistics, machine learning, data analysis, data mining, or data visualization, OR it focuses on programming, debugging, or performing routine operations within a statistical computing platform. If the latter, you could try the support links we maintain." – mdewey, whuber
If this question can be reworded to fit the rules in the help center, please edit the question.

0
$\begingroup$

use the 'mfx' package.

Below is the code that evaluates the marginal effects at mean and corresponding standard errors

probitmfx(formula = admit ~ gre + gpa, data = data, atmean = TRUE)

Marginal Effects:
     dF/dx  Std. Err.      z    P>|z|   
gre 0.00057907 0.00022282 2.5989 0.009353 **
gpa 0.16025934 0.06740084 2.3777 0.017421 * 
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 
$\endgroup$
  • $\begingroup$ Hi Subra, thank you so much for your help - is there any chance you could provide further details on how I can practically go about this for my data as I am having issues in getting it to work. I would be so grateful for any help. For example, does this work for a glmer with a random effect (as is my model) (for a start I am getting the error that function 'probitmfx' cannot be found but I have downloaded the mfx package). I should probably also mention that 'name.origin' is a dummy variable. $\endgroup$ – Lola2000 Aug 17 '15 at 16:45
  • $\begingroup$ I can get that to work but what is is supposed to show..? $\endgroup$ – Lola2000 Aug 17 '15 at 18:11
  • $\begingroup$ Can you get that to work? it did not work at my end. However, if you used allEffects(m1), it should give you the marginal effects at different levels of the predictor variable. ** also found out that 'mfx' does not work for glmer $\endgroup$ – subra Aug 17 '15 at 18:14
  • $\begingroup$ Well I also need to know significance and the standard errors ? Do you know a way to get these for marginal effects? And the marginal effects I get from that formula are very small (much smaller than what I am expecting) $\endgroup$ – Lola2000 Aug 17 '15 at 18:34
  • $\begingroup$ Honestly, I am not sure on this 'effects' package, bcos its not defined for 'glmer'. May we should wait until someone else comes up with an asnwer $\endgroup$ – subra Aug 17 '15 at 18:47

Not the answer you're looking for? Browse other questions tagged or ask your own question.