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

From |
Renuka Metcalfe <rm18203@yahoo.co.uk> |

To |
statalist <statalist@hsphsun2.harvard.edu> |

Subject |
st: 2SLS estimation with possible scaling |

Date |
Wed, 20 Feb 2008 12:41:31 +0000 (GMT) |

Dear Statalisters I have estimated 2SLS regressions. The data is cross-section and the endogenous explanatory variable is binary, I have used . probit x x1 x2 x3 z where x = endogen. binary rhs variable I then took the predicted values and ran an -ivreg- with the predicted values. I find that the coefficient of the endogenous variable is 1.05 and is significant. I guess I could scale the endogenous variable without scaling anything as yet. I did this and I find the coefficient of the endogenous variable is now 0.105 and has the same significance, that is, it is significant. But I am not certain, if would be correct. I would be grateful, if anyone would let me know if just scaling the endogenous variable is sufficient. I am not certain if I should scale the dependent variable of the ivreg and also the instrument I used in the first stage regression and the predicted values if this is possible. Thank you __________________________________________________________ Sent from Yahoo! Mail - a smarter inbox http://uk.mail.yahoo.com * * 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/

- Prev by Date:
**st: stsplit according to percentiles of a variable** - Next by Date:
**st: RE: stsplit according to percentiles of a variable** - Previous by thread:
**st: stsplit according to percentiles of a variable** - Next by thread:
**RE: st: How can I obtain the name of a variable's value label programatically?** - Index(es):

© Copyright 1996–2016 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |