Bookmark and Share

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, is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Fwd: Ologit mfx question

From   Richard Williams <>
To, statalist <>
Subject   Re: st: Fwd: Ologit mfx question
Date   Mon, 18 Feb 2013 22:00:26 -0500

You don't show your code, but I bet you did something like

use "";, clear
ologit warm male

ologit is a multiple outcome command, and hence there are multiple marginal effects to compute. In this case the marginal effect tells that being male increases the likelihood of your being in category 1, which is consistent with what the coefficient estimates tell you. Instead, you ought to be doing something like

use "";, clear
ologit warm male
mfx, predict(outcome(#1))
mfx, predict(outcome(#2))
mfx, predict(outcome(#3))
mfx, predict(outcome(#4))

Basically, the results tell you that being male increases the likelihood of being in the lower 2 categories and decreases the likelihood of being in the upper 2, which is consistent with the negative coefficient for male in the ologit results, i.e. if you are less likely to be in one category you have to be more likely to be in another.

You can make life a little simpler by using either -margeff- or -mfx2-, both of which can be downloaded from SSC.

Since you are using mfx, I assume you are using Stata 10 or lower. But if you are using Stata 11 or 12, the -margins- command is superior to -mfx-. But you still need to specify multiple outcomes.

At 09:41 PM 2/18/2013, ä¿?å?©,ä¿? Junli-Yu wrote:
Dear Statalister,

    The dependent var is Order value(0-1-2-3) ,and my key independent var is
dummy var(0-1). I regress my dataset by ologit that shows the coeffience is
positive significant relation, but by mfx shows the margin effect is
negative significant relation.

Whether it is wrong use mfx command. Because I recode my dependent var that
change to dummy variable (0-1),both the coeffience and margin effect are
same relation.
*   For searches and help try:

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

*   For searches and help try:

© Copyright 1996–2016 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index