Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: interpretation for negative and positive slope combination of interaction term
From
[email protected]
To
[email protected]
Subject
Re: st: interpretation for negative and positive slope combination of interaction term
Date
Thu, 9 May 2013 19:51:43 +0000
Sent from my BlackBerry® smartphone provided by Airtel Nigeria.
-----Original Message-----
From: Nahla Betelmal <[email protected]>
Sender: [email protected]
Date: Thu, 9 May 2013 20:51:21
To: <[email protected]>
Reply-To: [email protected]
Subject: Re: st: interpretation for negative and positive slope combination of
interaction term
Thanks for the reply David, but I think there is something not quite
right. If you check this file, p.133 figure 7.8
http://www.sagepub.com/upm-data/21120_Chapter_7.pdf
You will notice that in order to get the slope for the group with
Dummy= 1 (overconfident manager in my case), we should add the
coefficient of beta and gama ( MV and OC*MV in my case) , and to get
the intercept for those managers we should add the alpha and y ( my
model intercept and the coefficient of OC )
I found the same in other files, the problem is that all the examples
they provide both beta and gama are positive which makes the addition
and interpretation process easy.
my question is how to add and interpret when one is positive and the
other is negative . As the effect of MV on realistic managers is -
0.0566 and on overconfident managers is + 0.0596 , it moves from
negative to positive 0.003 (again the interaction is significant
although at 10 level). how many times MV effect overconfident managers
more than other managers ?
I would really appreciate help in that
Many thanks
Nahla
On 9 May 2013 19:36, David Crow <[email protected]> wrote:
> Dear Nahla-
>
> You're on the right track, but not quite right. I find that it's
> helpful to think of the meaning of each coefficient. Let's boil your
> model down to just the two variables (Market Value, MV, and
> Overconfident Managers, OC), their interaction, and an intercept:
>
> y = B0 + B1*(MV) + B2*(OC) + B3*(MV*OC) + u
> and yhat = B0 + B1*(MV) + B2*(OC) + B3*(MV*OC)
>
> Since OC is an indicator variable (overconfident = 1), when OC=0--that
> is, for non-overconfident, or "realistic" managers", yhat is simply B0
> + B1*(MV) and the effect of market value is given by B1. However,
> when OC=1--that is, for overconfident managers--yhat is
> B0+B1*MV+B2*OC+B3*MV*OC. Since OC=1, this simplifies to
> B0+B1*MV+B2+B3*MV and the effect of MV is given by B1+B2+B3
>
> Your calculation (-0.0566241 + 0.0596146= 0.003) leaves out the term
> B2, the coefficient for OC. So, the correct slopes are:
>
> OC=0: -0.0566241
> OC=1: -0.0566241 + -.1040174 + 0.0596146 = -.1010269.
>
> In this case, the effects of market value appear to attenuate the
> effects of overconfidence.
>
> Hope this helps.
>
> Best,
> David
>
> On Thu, May 9, 2013 at 8:20 AM, Nahla Betelmal <[email protected]> wrote:
>>
>> Dear Statalist,
>>
>>
>> As you can see below, I have a interaction term between OC (dummy =1
>> for overconfidence) and MV (continuous variable for market value). The
>> interaction term is positive and significant. I want to calculate the
>> slope against MB for overconfident managers which should be the
>> coefficient of MV plus the coefficient of OC*MV.
>> I am confused how to get this figure because MV is negative and OC*MV
>> is positive. So, Should it be -0.0566241 + 0.0596146= 0.003? if this
>> is true how can I interpret how many times the effect of MV is larger
>> for overconfident managers?? 0.003/0.0566.
>>
>> I am really confused and I highly appreciate your help please
>>
>>
>>
>>
>> Linear regression Number of obs = 49
>> F( 10, 38) = 3.23
>> Prob > F = 0.0043
>> R-squared = 0.4385
>> Root MSE = .08529
>>
>> ------------------------------------------------------------------------------
>> | Robust
>> earnings managment| Coef. Std. Err. t P>|t| [95%
>> Conf. Interval]
>> -------------+----------------------------------------------------------------
>> var1 | .0081153 .0058432 1.39 0.173 -.0037137 .0199443
>> MV | -.0566241 .0353602 -1.60 0.118 -.128207 .0149588
>> var3| .1992782 .093338 2.14 0.039 .0103252 .3882312
>> var4 | -.0040891 .0109331 -0.37 0.710 -.0262219 .0180437
>> var5 | .0817256 .1169071 0.70 0.489 -.1549405 .3183917
>> var6 | .0291373 .026944 1.08 0.286 -.0254079 .0836825
>> var7 | -.0646094 .0320074 -2.02 0.051 -.129405 .0001863
>> var8 | -.0867868 .0311875 -2.78 0.008 -.1499227 -.0236509
>> OC| -.1040174 .0556577 -1.87 0.069 -.2166906 .0086558
>> OC*MV | .0596146 .0324333 1.84 0.074 -.0060433 .1252724
>> _cons | .1643745 .0994735 1.65 0.107 -.0369991 .365748
>>
>>
>> many Thanks
>>
>> Nahla Betelmal
>> *
>> * For searches and help try:
>> * http://www.stata.com/help.cgi?search
>> * http://www.stata.com/support/faqs/resources/statalist-faq/
>> * http://www.ats.ucla.edu/stat/stata/
>
>
>
>
> --
> Web site for México, las Américas y el Mundo:
> http://mexicoyelmundo.cide.edu/
>
> ====================================
> David Crow
> Profesor-Investigador/Assistant Professor
> División de Estudios Internacionales
> Carretera México-Toluca 3655
> Col. Lomas de Santa Fe 01210 México, D.F.
> Tel.: 5727-9800, ext. 2152
> Fax: 5727-9872
> ====================================
>
> Conmutador: 5727-98-00 Lada sin costo: 01 800 021 2433 (CIDE) |©
>
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/faqs/resources/statalist-faq/
> * http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/