[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: graph question |

Date |
Thu, 15 May 2008 15:12:13 +0100 |

Good question. My answer: you need a different data structure first. Warning: this destroys your data. Assume variables v1-v7. Then consider gen id = _n reshape long v, i(id) contract v _j , zero graph hbar (asis) _freq , over(_j) over(v) You may do need to do some renaming to get the names in a suitable state for -reshape-. I used the -hbar- choice because for your problem with 28 bars, legibility will be an issue with -graph bar-. A hack at a slightly more general program that leaves your data intact. *! 1.0.0 NJC Statalist 15 May 2008 program multibar version 8 syntax varlist(numeric) [if] [in] [, VALues *] marksample touse, novarlist quietly { count if `touse' if r(N) == 0 error 2000 preserve keep if `touse' keep `varlist' tempname stub lbl local i = 1 foreach v of local varlist { gen `stub'`i' = `v' label def `lbl' `i' "`v'", add local ++i } tempvar id gen `id' = _n reshape long `stub', i(`id') label val _j `lbl' contract `stub' _j, zero } if "`values'" != "" { graph hbar (asis) _freq, over(_j) over(`stub') `options' } else graph hbar (asis) _freq, over(`stub') over(_j) `options' end That is a bit awkward, as in practice you might want to reach inside and add suboptions to the -over()- options. In many ways, the best strategy is 1. New structure first. 2. Then -graph-. Hoping to do it all via one -graph- command is probably hoping for too much. Nick n.j.cox@durham.ac.uk Tunga Kantarci I have a graphing question: I have 7 variables. All the variables take values from 1 to 4. I would like to produce a single graph on which the frequencies are categorized for the seven variables, for each of the 4 values. That is, for value one depict a frequency bar for the first variable depict a frequency bar for the second variable next to it depict a frequency bar for the third variable next to it for value two depict a frequency bar for the first variable depict a frequency bar for the second variable next to it depict a frequency bar for the third variable next to it for value three depict a frequency bar for the first variable depict a frequency bar for the second variable next to it depict a frequency bar for the third variable next to it for value four depict a frequency bar for the first variable depict a frequency bar for the second variable next to it depict a frequency bar for the third variable next to it Here the categorization is over "the values the seven variables take", and it is not over "a variable". I tried several commands of bar graph but cannot really get what I want. * * 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: SUM question** - Next by Date:
**RE: st: creating a variable using -if-? programming needed?** - Previous by thread:
**st: Re: Re: GMM or 2SLS with cluster adjustment or others** - Next by thread:
**st: a second test.** - Index(es):

© Copyright 1996–2014 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |