Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

# st: economic interpretation mfx!!

 From "Augusto Pallavicini" To Subject st: economic interpretation mfx!! Date Fri, 27 Jul 2012 19:53:36 +0100

Hi all, I need help with the interpretation of the output of the marginal effects after tobit... I know this is conditional...
```
here is what I think...

```
One more level of education increases the expected amount willing to pay by 61.2% and one more level of income by 29.2%, ceteris paribus. Then, having a positive opinion about wind energy (dummy opinion2) raises the expected amount willing to pay by 208.7% whereas being concern about the environment (dummy statement2) does it by 148.2%, ceteris paribus
```
```
Are these conclusions correct? Do I need to do something else with the dummies to find out the mg effect (because I think it's wrong)?
```

Below is the output

mfx compute, predict(e(0,.))

Marginal effects after tobit
y  = E(pay|pay>0) (predict, e(0,.))
=   4.332974
------------------------------------------------------------------------------
```
variable | dy/dx Std. Err. z P>|z| [ 95% C.I. ] X
```---------+--------------------------------------------------------------------
```
educat~n | .612348 .33726 1.82 0.069 -.048671 1.27337 2.84058 income | .2918185 .17118 1.70 0.088 -.043688 .627325 3.50725 gender*| -.1665404 .63723 -0.26 0.794 -1.4155 1.08242 .521739 age | -.0552524 .40434 -0.14 0.891 -.847737 .737232 3.52174 coastal*| -.8215083 .63314 -1.30 0.194 -2.06245 .419432 .623188 bill | .0053813 .01494 0.36 0.719 -.023894 .034657 55.0435 children*| .5257885 .83808 0.63 0.530 -1.11681 2.16839 .623188 opinio~2*| 2.087348 .6481 3.22 0.001 .817098 3.3576 .869565 statem~2*| 1.481657 .62187 2.38 0.017 .262813 2.7005 .797101
```------------------------------------------------------------------------------
(*) dy/dx is for discrete change of dummy variable from 0 to 1

```
AP
```*
*   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/
```