[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

From |
Berthold Hoppe <berthold.hoppe@charite.de> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
st: tobit or anything else? |

Date |
Sat, 12 Apr 2008 09:48:00 +0200 |

Dear Statalist users!

I try to analyze the relation between two continuous biological markers (y, x) in a study population, where -y- depends on -x-. Other data indicate that this relation is significantly influenced by a genotype z (two outcomes: 0, 1). The hypothesis is that the regression coefficients (y, x) differ between z=0 and z=1.

As -y- in censored (due to lower detection limit of the measurements) -tobit- instead of -reg- seems to be adequate.

Would the following analysis be correct to test my hypothesis?

xi3: tobit y x@z, ll

test x@z==0 = x@z==1

Does anyone know, how to handle this question.

Have a nice weekend,

Berthold

*

* For searches and help try:

* http://www.stata.com/support/faqs/res/findit.html

* http://www.stata.com/support/statalist/faq

* http://www.ats.ucla.edu/stat/stata/

**Follow-Ups**:**st: RE: tobit or anything else?***From:*"Verkuilen, Jay" <JVerkuilen@gc.cuny.edu>

- Prev by Date:
**st: Inverse Hyperbolic Transformation with 3SLS** - Next by Date:
**st: Re: list [if] [in]** - Previous by thread:
**st: Inverse Hyperbolic Transformation with 3SLS** - Next by thread:
**st: RE: tobit or anything else?** - Index(es):

© Copyright 1996–2015 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |