Notice: On March 31, it was **announced** that Statalist is moving from an email list to a **forum**. The old list will shut down on April 23, and its replacement, **statalist.org** is already up and running.

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

From |
"Carlo Lazzaro" <carlo.lazzaro@tin.it> |

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

Subject |
R: st: error 506 (matrix not positive definite) when bootstrapping eigenvalue after one-way manova |

Date |
Wed, 15 Jun 2011 07:40:12 +0200 |

Dear Sergiy, thanks a lot for your appreciated insights about this (for me) still weird issue. Yes, I have taken a look at the printed version of Stata manual that came with the 9.2 release, but I did not find out any useful diagnostic advice for tracking down what corrupted the matrix. This strikes me even more since foreign variable was not affected by any transformation/manipulation. Kindest Regards, Carlo -----Messaggio originale----- Da: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] Per conto di Sergiy Radyakin Inviato: mercoledì 15 giugno 2011 2.00 A: statalist@hsphsun2.harvard.edu Oggetto: Re: st: error 506 (matrix not positive definite) when bootstrapping eigenvalue after one-way manova Hello Carlo, below is a somewhat more compact code that still exhibits the same problem as you have. Note: 1) the use of the version statement 2) sysuse is used here instead of use not to bother about the path 3) no magic numbers in the code, constants are determined from data 4) _manova is called internally by manova. The error 506 is related to matrix not being positive definite. You may want to read the formulas in the Stata documentation (PDF or printed, not help) for manova to figure out exactly why this is a requirement. Best, Sergiy Radyakin //-------------------------------------------------------------------------- -------------- version 9 clear sysuse auto // THIS WORKS _manova price mpg=i.foreign [], // TRANSFORM VARIABLES sum price if foreign local foreign_price=r(mean) sum price if foreign==0, meanonly replace price=price-r(mean)+`foreign_price' if foreign==0 sum price if foreign==1, meanonly replace price=price-r(mean)+`foreign_price' if foreign==1 sum mpg if foreign local foreign_mpg=r(mean) sum mpg if foreign==0, meanonly replace mpg= mpg-r(mean)+`foreign_mpg' if foreign==0 sum mpg if foreign==1, meanonly replace mpg= mpg-r(mean)+`foreign_mpg' if foreign==1 // TRY AGAIN, THIS DOES NOT WORK _manova price mpg=i.foreign [], //-------------------------------------------------------------------------- -------------- END -------------- * * 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:
**Re: st: Using sampling/probability weights for mixed design ANOVA in STATA** - Next by Date:
**Re: st: long table** - Previous by thread:
**Re: st: Using sampling/probability weights for mixed design ANOVA in STATA** - Next by thread:
**st: Re: statalist-digest V4 #4184** - Index(es):