From tburch@fas.harvard.edu Mon May 26 21:51:42 2003 From: tburch@fas.harvard.edu To: gov2001@lists.gking.harvard.edu Subject: [gov2001-l] Ordered Logit Date: Mon, 26 May 2003 16:51:42 -0400 Message-ID: <008901c323c8$9c89e520$c8bb6780@dhcp186172> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="===============0794997963740356648==" --===============0794997963740356648== Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit This is a multi-part message in MIME format. ------=_NextPart_000_0086_01C323A7.1559C0A0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable How does one interpret ordered logit coefficients? I remember from = class that regular probit coefficients mean something like the fraction = of a standard deviation the underlying coefficient will = increase/decrease with a one unit increase in x. Is there a similar way = of understanding ordered logit? Traci ------=_NextPart_000_0086_01C323A7.1559C0A0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
How does one interpret ordered logit=20 coefficients?  I remember from class that  regular probit = coefficients mean something like the fraction of a standard deviation = the=20 underlying coefficient will increase/decrease with a one unit increase = in=20 x.  Is there a similar way of understanding ordered = logit?
 
Traci
------=_NextPart_000_0086_01C323A7.1559C0A0-- --===============0794997963740356648==-- From king@harvard.edu Tue Feb 9 22:44:19 2021 From: Gary King To: gov2001@lists.gking.harvard.edu Subject: [gov2001-l] Ordered Logit Date: Mon, 26 May 2003 17:01:16 -0400 Message-ID: In-Reply-To: <008901c323c8$9c89e520$c8bb6780@dhcp186172> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="===============5056095838454911149==" --===============5056095838454911149== Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit one way is as a linear regression with the outcome variable being the unobserved continuous variable, Y* that has stadnard deviation 1. a better way is to do simulations for each of the categories, given some specific values of the X's. Gary On Mon, 26 May 2003, Traci Burch wrote: > How does one interpret ordered logit coefficients? I remember from > class that regular probit coefficients mean something like the fraction > of a standard deviation the underlying coefficient will > increase/decrease with a one unit increase in x. Is there a similar way > of understanding ordered logit? > > Traci > --===============5056095838454911149==--