st: =?utf-8?B?UkU6IFJFOiBSRTogUkU6IFJFOiBSRTogUkU6IFJFOiA=?==?utf-8?B?UkU6IFJFOiBSRTogc3Q6IMK0w7DCuMK0OiBzdDogSG93IHRv?==?utf-8?B?IGNob29zZSBhIHByb3BlciBtb2RlbCBpZiB0aGUgZGU=?==?utf-8?B?cGVuZGVudCB2YXJpYWJsZSBpcyB3aXRoaW4gYm91bmQ=?==?utf-8?B?cz8=?=

["Cheng, Xiaoqiang" <Xiaoqiang.Cheng@econ.kuleuven.be>]

Thank you very much for your advice.

I will try.

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 Nick Cox
Sent: 2006年10月2日 16:47
To: statalist@hsphsun2.harvard.edu
Subject: st: RE: RE: RE: RE: RE: RE: RE: RE: RE: RE: st: ´ð¸´: st: How to choose a proper model if the dependent variable is within bounds?

My best advice to you is try various changes of units yourself 
and note the results obtained. That way you will better understand 
what -mfx- does and be able to work out what is relevant for
your problem. You won't make much progress with Stata if you expect
Statalist members to hold your hand every tiny step along the way. 

Nick 
n.j.cox@durham.ac.uk 

Cheng, Xiaoqiang
> 
> Thank you very much.
> 
> Why I ask so because Maarten suggests rescale the variable x 
> before running -glm- if the approximation of slope is 
> exceeding unit. What I asked is can we simply do it another 
> way around: i.e., multiply the slope by a smaller change of x 
> such as 1/100 unit of x rather than 1 unit of x, and then we 
> can explain like by 1/100 unit change of x will increase a 
> certain level of y, which will not exceed 1. 
> 
> I don’t know if I make myself clear this time:)?

Nick Cox
 
> I don't think guessing is needed here, as this is basic 
> trigonometry, but yes, 
> vertical change = horizontal change X slope. 
 
Cheng, Xiaoqiang
  
> > To make it clear, let's forget what I have said.
> > 
> > What I want to know is, since dy/dx is the approximate slope 
> > and it should be explained as how much y will change respect 
> > to one unit change of x. In this case, I guess it will be 
> > right if we devide the value of the slope by 100 in order to 
> > know how much y will change respect to 1/100 unit change of x.

*
*   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/