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

From |
"Nick Cox" <n.j.cox@durham.ac.uk> |

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

Subject |
st: RE: re: Stata's row and column labels for matrices |

Date |
Mon, 28 Mar 2005 17:16:18 +0100 |

If you don't like Stata's default names r1, r2, ..., c1, c2, ..., and/or you want more informative row and column names, then you can supply them yourself. That is, the onus is on you to supply names as desired. The suggestion that Stata's default row and column labelling should echo the names of the matrices is interesting, but surely undesirable in practice. The limit on matrix name length and matrix row/column name length appears to be 32 characters, at least in Stata 8 SE. A fairly long matrix name and perhaps fairly long row or column name might well be desirable in some circumstances, but the consequences of using the matrix name in general to label rows and columns would often be matrix displays difficult to read, as a few experiments with names and -matrix list- make clear. The present default r1, r2, ..., c1, c2, ... is a little ugly but by contrast simple, clear and practical. Nick n.j.cox@durham.ac.uk P.S. please don't send HTML to the list. Elmer Villanueva Thanks for this. Your right, of course. I feel rather stupid having raised it in the first place. I've got a simple brain and it seemed to me that if the resulting matrices were labelled r1, r2... and c1, c2..., even if they weren't really rows and columns, it could easily be misinterpreted by people like me, especially if the matrix calculations are complex and there aren't any suitable markers. But, as you say, they're mere names and I'm sure Stata's got it right all along. Anyway, matrices don't really need column or row names at all. However, if rows and columns are named, shouldn't they be labelled more informatively? Perhaps they can even consider adopting your notation automatically so that matrices created by hand aren't unimaginatively named. I guess I was thinking that output should be informative or it should be dropped. You've showed, I suppose, that the calculation involving eight hand-entered matrices A-H producing a single matrix with names r1, r2... for both rows and columns simply means that the results were derived from rows of one matrix and the rows of another. Which ones? Irrelevant. As you say, there's really nothing to it and Stata's got it right. * * 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:
**RE: st: Stata Journal 5(1) 2005** - Next by Date:
**st: RE: levinlin vs. ipshin: lag length with local macro** - Previous by thread:
**Re: st: Stata Journal 5(1) 2005** - Next by thread:
**st: RE: levinlin vs. ipshin: lag length with local macro** - Index(es):

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