Re: st: interpretation for negative and positive slope combination of interaction term

[Nahla Betelmal <nahlaib@gmail.com>]

Thank you David and Jay for your help and the link. Sorry for the
delay reply as I was not feeling well.

the dependent variable is expressed as ratio (abnormal accrual scaled
by total assets), and the MV is expressed as ratio as well (market
value of assets scaled by total assets). So I think I should have
said:
 excessively overconfident bidders have a greater incentive to manage
earnings via accruals, approximately 0.0596 percent  more than rational
 bidders for each one percent increase, for MV.

I have also tried the above the average MV ( dummy = 1 if the bidder
MV is higher than the industry). The coefficent for MV above the
industry is -0.0592, and the interaction term for both dummies (OC*MV
above industry)= 0.0782, both are significant. Will that be expressed
as ratio or odds??

 I will also try increase the sample size, use other proxy for MV, and
apply log ,  poisson vce(robust)

 By the way the book is John Fox (1997), Applied Regression Analysis,
Linear Models, and Related Methods, Sage
 Publications

Thanks again, and please any more comments are more than welcome,

 Nahla
*
*   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/