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

From |
Richard Williams <[email protected]> |

To |
[email protected] |

Subject |
Re: st: Inconsistent Result of generating variables |

Date |
Fri, 09 Apr 2004 21:08:03 -0500 |

At 09:22 PM 4/9/2004 -0400, Jitian Sheu wrote:

I hope you mean you got the same results as I did, not the same incorrect results you got earlier! Anyway, here is the Stata FAQ I was thinking of - "I coded 'a'2 and got the wrong sign. Why?" SeeHello, Thank you very much. Based on your explaination, I try to use: (`b21')^2 instead of using `b21'^2. Stata creates the same results.

http://www.stata.com/support/faqs/lang/macpow.html

The trick is that when you say,

local b21=-0.25

You naturally assume that Stata will think `b21' means "negative .25". But, Stata actually interprets it as "minus .25". Hence, with `b21'^2, exponentiation gets done first and then subtraction, leaving you with -.0625 rather than +.0625.

The FAQ suggests two workarounds:

* Use scalars instead of local macros, e.g.

scalar b21 = -.25

* Or, bind the macro in parentheses when taking powers, e.g

(`b21')^2

*

* 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**:**Re: st: Inconsistent Result of generating variables***From:*"Jitian Sheu" <[email protected]>

**References**:**Re: st: RE: tabodds.ado***From:*[email protected]

**Re: st: Inconsistent Result of generating variables***From:*Richard Williams <[email protected]>

**Re: st: Inconsistent Result of generating variables***From:*"Jitian Sheu" <[email protected]>

- Prev by Date:
**RE: st: Residuals in Logistic Regression** - Next by Date:
**Re: st: Inconsistent Result of generating variables** - Previous by thread:
**Re: st: Inconsistent Result of generating variables** - Next by thread:
**Re: st: Inconsistent Result of generating variables** - Index(es):

© Copyright 1996–2024 StataCorp LLC | Terms of use | Privacy | Contact us | What's new | Site index |