[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
itchiftc@rumms.uni-mannheim.de |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: re: one-sided p-value using test x1=x2 |

Date |
Thu, 01 Oct 2009 21:35:03 +0200 |

Thanks! This helped me a lot! Zitat von Kit Baum <kitbaum@mac.com>:

<> Does it mean, that if I have two negative coefficients and want to test whether the one (which has the more negative point estimate) is significantly more negative then the other, I can divide the p-value by two and reject (or resepctively not reject) the null-hypothesis? Yes. If the null is that the coefficients are equal (or, in one-tailed terms, h0: \beta_1 >= \beta_2) then what you need to reject on a one-tailed basis is that \beta_1 << \beta_2 in terms of the point estimates. The p-val for that test is one-half the two-sided p-val reported for a test of equality of the coefficients. Kit Baum kitbaum@me.com * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

- Prev by Date:
**st: weird margins** - Next by Date:
**Re: st: Why include leads (in a model with lags)?** - Previous by thread:
**st: weird margins** - Next by thread:
**Re: st: Why include leads (in a model with lags)?** - Index(es):

© Copyright 1996–2014 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |