Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.

# Re: st: st: Interpreting mfx & margeff Coefficients from GLM

 From Richard Williams To statalist@hsphsun2.harvard.edu, statalist@hsphsun2.harvard.edu Subject Re: st: st: Interpreting mfx & margeff Coefficients from GLM Date Tue, 03 Apr 2012 09:29:59 -0500

```At 04:01 PM 4/2/2012, Clifton Chow wrote:
```
```Stata listers,

```
I ran a glm log link (family gamma from modified park's test) on my dataset in which the dependent variables is hours worked. To interpret the coefficients in the model, I transformed it via mfx and margeff. I have a couple of questions about interpreting the coefficients after transformation;
```
```
1. Is mfx mainly appropriate for indicator explanatory variables and margeff for continous ones? 2. Can I interpret the results of the mfx coefficients in the raw scale (i.e. hours)?
```3. What does the column in the far right X referred to in the mfx output?
```
```
```
Most people are going to want to use the -margins- command rather than the older -mfx- or -margeff-. The far right column in -mfx- is the mean of the X variable. I am not sure what is happening with this particular combo of link and family, but an intro to -margins- is at http://www.nd.edu/~rwilliam/stats/Margins01.pdf .
```
```
```Below is my partial mfx output from GLM:

y  = Predicted mean hours (predict)
=  8.7005381
------------------------------------------------------------------------------------------------
```
variable | dy/dx Std. Err. z P>|z| [ 95% C.I. ] X
```---------+--------------------------------------------------------------------------------------
```
X1*| 8.891796 1.83527 4.84 0.000 5.29474 12.4889 .237385 X2*| -5.151794 1.41499 -3.64 0.000 -7.92513 -2.37846 .519545
```(*) dy/dx is for discrete change of dummy variable from 0 to 1

```
4. For margeff, to interpret these in hours, should I exp(margeff coefficient)?
```

Average partial effects after glm
y  = log(hours)

------------------------------------------------------------------------------
```
variable | Coef. Std. Err. z P>|z| [95% Conf. Interval]
```-------------+----------------------------------------------------------------
```
X1 | 16.5519 3.9611 4.18 0.000 8.788287 24.31551
``` X2         |  -7.379914   1.553438    -4.75   0.000     -10.4246   -4.335232

Thanks
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```
```
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW:    http://www.nd.edu/~rwilliam

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```