LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (October 2002)Back to main SPSSX-L pageJoin or leave SPSSX-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Thu, 24 Oct 2002 14:41:32 -0400
Reply-To:     Richard Ristow <wrristow@mindspring.com>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Richard Ristow <wrristow@mindspring.com>
Subject:      Re: regression models
Comments: To: "kalanda@liverpool.ac.uk" <bkalanda@yahoo.com>
In-Reply-To:  <20021024083852.74763.qmail@web40305.mail.yahoo.com>
Content-Type: text/plain; charset="us-ascii"; format=flowed

At 01:38 AM 10/24/2002 -0700, Boniface Kalanda<kalanda@liverpool.ac.uk> wrote:

>I have two regression models: > y= a + bx > y= c+ dx >How do i test the hypotheses that > a is equal to c AND > b is equal to d >I would like to do that either through code or menus.

Those two models are structurally identical, so the question makes sense only if they are estimated on different data (different variables, or the same quantities measured in different instances, for x and y).

This is a "homogeneity of regression" problem, then: do the regression parameters differ in two sets of data?

You need to construct a data set containing both the cases over which you are estimating the first model and those over which you are estimating the second, AND a variable I will call "C_MODEL", which is 1 in those cases belonging to the second model and 0 in those cases belonging to the first.

Then, estimate the model

y = INT + SLOPE*x + INT_DIF*C_MODEL + SLOPE_DF*(C_MODEL*x)

(To do this, you have to compute a variable whose value is C_MODEL*x.)

Then, the test you want is that the two parameters INT_DIF and SLOPE_DF do *not* contribute significantly to the model.


Back to: Top of message | Previous page | Main SPSSX-L page