# st: =?utf-8?B?UkU6IFJFOiBSRTogUkU6IFJFOiBzdDogwrTDsMK4wrQ6IHN0?==?utf-8?B?OiBIb3cgdG8gY2hvb3NlIGEgcHJvcGVyIG1vZGVsIGk=?==?utf-8?B?ZiB0aGUgZGVwZW5kZW50IHZhcmlhYmxlIGlzIHdpdGg=?==?utf-8?B?aW4gYm91bmRzPw==?=

 From "Cheng, Xiaoqiang" To Subject st: =?utf-8?B?UkU6IFJFOiBSRTogUkU6IFJFOiBzdDogwrTDsMK4wrQ6IHN0?==?utf-8?B?OiBIb3cgdG8gY2hvb3NlIGEgcHJvcGVyIG1vZGVsIGk=?==?utf-8?B?ZiB0aGUgZGVwZW5kZW50IHZhcmlhYmxlIGlzIHdpdGg=?==?utf-8?B?aW4gYm91bmRzPw==?= Date Mon, 2 Oct 2006 15:38:35 +0200

```Thank you very much.

Can we think an alternative of your previous suggestion about the prediction of dy/dx when it exceeds unit?

One example will be: if the prediction exceeds unit and x is counted in 100s (one unit of x in sample data means 100 in reality. For instance, x stands for kilometres. In reality, the distance from Brussels to Leuven is 20 km, but in our sample we count the distance as 0.2), can we interpret the dy/dx by, lets say, one unit of x increase in reality, the y will increase (dy/dx)/100?

Xiaoqiang Cheng

University of Leuven
Tel  +32 16 326853
Fax  +32 16 326796
Mail  Xiaoqiang Cheng
Center for Economic Studies
University of Leuven
Naamsestraat 69
Leuven,  Belgium
B3000
Url www.econ.kuleuven.be/xiaoqiang.cheng

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Maarten Buis
Sent: 2006年10月2日 15:05
To: statalist@hsphsun2.harvard.edu
Subject: st: RE: RE: RE: RE: st: ´ð¸´: st: How to choose a proper model if the dependent variable is within bounds?

Xiaoqiang Cheng
These questions are answered by the helpfileof -mfx-:
-mfx- will also work for -probit-, -logit- and
-tobit-. In case of -probit-, and -logit- I actually
prefer -prchange-, which is part of the -spost-
package (see -findit spost-). For -tobit- there also
exist the -dtobit- user written command (see:
-findit dtobit-). You can specify the values of x at which
you want to calculate the marginal effect  with the at()
option, see -help mfx-.
HTH,
Maarten

-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

http://home.fsw.vu.nl/m.buis/
-----------------------------------------

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu]On Behalf Of Cheng, Xiaoqiang
Sent: maandag 2 oktober 2006 14:46
To: statalist@hsphsun2.harvard.edu
Subject: st: RE: RE: RE: st: ´ð¸´: st: How to choose a proper model if the dependent variable is within bounds?
how to specify a certain value of x if I want to predict the dy/dx using -mfx- for the certain value of x? And is there any similar postestimation command like -mfx- for -probit-, -logit- and -tobit-?

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```

• References: