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: indicator variable and interaction term different signs but both significant
From
Richard Williams <[email protected]>
To
[email protected], [email protected]
Subject
Re: st: indicator variable and interaction term different signs but both significant
Date
Sat, 06 Apr 2013 23:23:15 -0500
Check the handout I sent before
(http://www3.nd.edu/~rwilliam/xsoc63993/l53.pdf) and see if that
answers most of your Qs. Briefly, the differing signs don't matter.
My claim that the effect of MV is 3 times larger for overconfident
managers stems from the fact that the effect of MV was about .005 in
your earlier results and the interaction was around .010, which means
that the effect of MV is about .015 for the overconfident.
You may also want to look at
<http://www3.nd.edu/~rwilliam/xsoc63993/l51.pdf>http://www3.nd.edu/~rwilliam/xsoc63993/l51.pdf
See Especially p. 13, as I think it addresses your biggest source of
confusion. Indeed, I added the graphic on that page several years ago
precisely because students got confused over things like this.
At 06:53 PM 4/6/2013, Nahla Betelmal wrote:
Hi Richard,
Thank you for the help and for the file.
let me get this straight, if MV=0 , overconfident managers would
manage earnings LESS than other managers (accept as it is
statistically significant). However, When MV has other values the
overconfident manage might manage earnings more or less than others.
I understand that interpretation of OC_D has no practical meaning as
MV cant be zero, however, I wonder if the interpretation of the
interaction term (OC_MV) depends on the sign of OC_D. As you can see
they have different signs.
Does the fact OC_D has a negative sign and OC_MV has a positive one ,
mean : a) overconfident managers will manage earnings more when MV is
high, or b) overconfident mangers will manage earnings less especially
when MV is high!!
Also, can you explain what do you mean by "To me the critical thing
seems to be that the effect of MV is about 3 times as large for
overconfident managers as it is for regular managers" I did not get
that.
The other variables have theoretical background, however, I drooped
them to see what happens.
Linear regression Number of obs = 56
F( 3,
52) = 2.90
Prob >
F = 0.0433
R-squared
= 0.1116
Root
MSE = .09448
------------------------------------------------------------------------------
| Robust
Earnings Mgt | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
MV | .000584 .0015479 0.38 0.707 -.0025221 .0036901
OC_D | -.0728009 .0320739 -2.27 0.027 -.137162 -.0084398
OC_MV | .012836 .0048402 2.65 0.011 .0031235 .0225485
_cons | .0156561 .0125722 1.25 0.219 -.0095719 .0408841
Thank you again
Nahla
On 7 April 2013 00:32, Richard Williams <[email protected]> wrote:
> The significance of OC_D is not any big deal, or at least need
not be. It is
> saying that, when MV = 0, the predicted difference between the two types of
> managers is statistically significant. At some other values of MV
it may not
> be statistically significant. Since you say MV is continuous
positive, 0 is,
> at best, the lowest possible value for MV, and in practice the lowest value
> may be much higher than 0.
>
> To give another example, suppose that instead of MV the variable
was weight.
> At 0 pounds the difference between two groups may be significant.
But nobody
> weighs 0 pounds so it isn't a useful comparison. Center weight so that 0 =
> average weight, and the comparison becomes more useful, i.e. the
coefficient
> for OC_D would be the expected different between the two types of managers
> who were both of average weight.
>
> If you wanted to, you could compute the predicted difference between two
> managers who both weighed a negative millions pounds or a positive million
> pounds. The coefficients might be wildly significant but hardly helpful.
>
>
> At 06:45 PM 4/6/2013, Nahla Betelmal wrote:
>>
>> Thanks Anthony for the reply. Actually Overconfidence indicator
>> variable takes value of 1 for overconfident managers, and value of
>> zero for rational managers. The market value variable is a continuous
>> positive variable.
>>
>> I am not sure that interpreting the indicator variable (OC_D) has
>> literal interpretation in the interaction model, I am confused due to
>> the significance ! otherwise I would only have focused on the
>> interaction term (i.e. overconfident managers manipulate earnings
>> upwardly when the market value of their firms is high) but I am not
>> sure if I got it right.
>>
>> Many thanks, and I am looking for others responses as well
>>
>> Nahla
>>
>>
>> On 6 April 2013 23:24, Anthony Fulginiti <[email protected]> wrote:
>> > Hi Nahla,
>> >
>> > I would recommend waiting for others on Statalist to respond to provide
>> > confirmation of my interpretation. However, my thoughts are
that this is
>> > suggesting that your main effect for overconfidence is suggesting that
>> > overconfident managers manipulate earnings less than other
managers (if that
>> > is the reference group) at market value 0. The interaction would then
>> > suggest that the effect of the overconfidence variable on earnings
>> > manipulation is increasingly greater at higher market values. I look
>> > forward to hearing other replies.
>> >
>> >
>> > Anthony
>> >
>> >
>> > On Apr 6, 2013, at 2:45 PM, Nahla Betelmal wrote:
>> >
>> >> Dear Statalist,
>> >>
>> >> I am having difficulty interpreting the results from OSL regression. I
>> >> am trying to see whether Overconfident managers manipulate earnings in
>> >> a certain context.
>> >>
>> >> The dependent variable earnings_Mgt is continuous The problem is that
>> >> the indicator variable for overconfidence (OC_D) is negative and
>> >> significant, while the interaction between the indicator variable and
>> >> market_value variable (OC_MV) is positive and significant. What does
>> >> that mean?
>> >> Does it mean that overconfident managers manipulate earnings less than
>> >> others (rational managers), but when the market value is high they
>> >> manipulate earnings more than rational managers do?
>> >>
>> >> Your help is highly appreciated,
>> >>
>> >> many thanks
>> >>
>> >> Nahla Betelmal
>> >>
>> >>
>> >> Linear regression Number of obs = 56
>> >> F( 8, 47) = 3.60
>> >> Prob > F = 0.0025
>> >> R-squared = 0.3719
>> >> Root MSE = .08355
>> >> Robust
>> >> earnings Mgt Coef. Std. Err. t P>t
>> >> [95% Conf. Interval]
>> >> size .0058268 .0092169 0.63 0.530 -.0127153
>> >> .0243689
>> >> leverage .0924198 .0724032 1.28 0.208 -.0532367
>> >> .2380763
>> >> MV .0046896 .0032752 1.43 0.159 -.0018993
>> >> .0112784
>> >> litigation .0310148 .0267527 1.16 0.252
>> >> -.0228048 .0848344
>> >> private_D -.0638102 .023056 -2.77 0.008 -.110193
>> >> -.0174275
>> >> same_D -.08197 .0273465 -3.00 0.004 -.136984
>> >> -.026956
>> >> OC_D -.0730767 .0288269 -2.54 0.015 -.131069
>> >> -.0150844
>> >> OC_MV .0105348 .0049493 2.13 0.039 .000578
>> >> .0204916
>> >> _cons .0381444 .0615391 0.62 0.538 -.0856564
>> >> .1619452
>> >> *
>> >> * 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/
>
>
> -------------------------------------------
> Richard Williams, Notre Dame Dept of Sociology
> OFFICE: (574)631-6668, (574)631-6463
> HOME: (574)289-5227
> EMAIL: [email protected]
> WWW: http://www.nd.edu/~rwilliam
>
> *
> * 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/
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME: (574)289-5227
EMAIL: [email protected]
WWW: http://www.nd.edu/~rwilliam
*
* 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/