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]

From |
David Hoaglin <dchoaglin@gmail.com> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: question about the interaction term |

Date |
Thu, 25 Apr 2013 14:11:23 -0400 |

Dear Vic, The output is very helpful. Thank you. The effects of frbc, homo, and frbc_homo must be interpreted as "after adjusting for the contributions of [the other predictors in the model]." With that in mind, none of those three predictors has a significant effect (at the .05 level). Thus, the effect (slope) of frbc does not differ between the homo=0 group and the homo=1 group. (I do not see where "frbc is significant in the testing.") It would be reasonable to remove frbc_homo from the model, to see whether frbc has a significant slope (overall) and whether homo is significant. If both of those variables are significant, then the same slope on frbc applies to both groups, and the two groups differ in level (i.e., intercept). David Hoaglin On Thu, Apr 25, 2013 at 11:21 AM, ZHVictor <victerzj2@hotmail.com> wrote: > Dear David, > Thank you!This is the output in stata > | Robust > | Coef. Std. Err. z P>|z| [95% Conf. Interval] > -------------+---------------------------------------------------------------- > performance | > frbc | -.8474035 .4923128 -1.72 0.085 -1.812319 .1175119 > frbc_homo | .3386407 .6860396 0.49 0.622 -1.005972 1.683254 > homo | .0977946 .4812832 0.20 0.839 -.8455031 1.041092 > board | .0076623 .0470874 0.16 0.871 -.0846274 .099952 > indep | -.5825246 .6177608 -0.94 0.346 -1.793314 .6282644 > audit | -.002093 .0926196 -0.02 0.982 -.1836241 .179438 > roa | 1.704518 1.494709 1.14 0.254 -1.225058 4.634093 > finance | .6123755 .430509 1.42 0.155 -.2314066 1.456158 > leverage | .0535289 .6785705 0.08 0.937 -1.276445 1.383503 > size | .0170652 .1050801 0.16 0.871 -.188888 .2230183 > mtb | .0545689 .027851 1.96 0.050 -.0000181 .109156 > inst_holding | .6830099 .6363314 1.07 0.283 -.5641768 1.930197 > _cons | -1.426625 1.042372 -1.37 0.171 -3.469637 .6163864 > test frbc+frbc_homo=0 > chi2( 1) = 1.87 > Prob > chi2 = 0.1718 > > frbc is a continuous variable and homo is a dummy variable. Basically, I want to see whether frbc has different effect in the homo=0 group vs homo=1 group. > The interaction term frbc_homo is insignificant but the frbc is significant in the testing "frbc+frbc_homo=0" > Thank you! * * 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/

**References**:**st: question about the interaction term***From:*ZHVictor <victerzj2@hotmail.com>

**Re: st: question about the interaction term***From:*David Hoaglin <dchoaglin@gmail.com>

**RE: st: question about the interaction term***From:*ZHVictor <victerzj2@hotmail.com>

**Re: st: question about the interaction term***From:*David Hoaglin <dchoaglin@gmail.com>

**RE: st: question about the interaction term***From:*ZHVictor <victerzj2@hotmail.com>

- Prev by Date:
**Re: st: How to normalize variables in stata** - Next by Date:
**RE: st: FW: vecrank yields "too many literals" error in small Stata** - Previous by thread:
**RE: st: question about the interaction term** - Next by thread:
**st: spatial ologit or spatial mlogit** - Index(es):