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

From |
SamL <saml@demog.berkeley.edu> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: drop variable on condition |

Date |
Mon, 21 Jul 2008 07:23:27 -0700 (PDT) |

The code below is helpful, but it is also an example of why some find stata a bit more challenging than some other packages. Anytime I want to do something conditional it *seems* to take an awful lot of code, often very complicated code. Really old SPSS allowed one to just write -do if- as a "pre-command" to almost anything (and -else if's- could be added if needed), and that made the following recode or generate or "whatever" conditional. Is there any way some such easy pre-command could be instituted in stata? Is it already available, and I just have failed to find it?

Any assistance is greatly appreciated.

Sam

On Mon, 21 Jul 2008, Andrea Bennett wrote:

Ah, thank you very much! Kind regards! On Jul 21, 2008, at 12:54 PM, Joseph Coveney wrote:Andrea Bennett wrote: A quick question related to dropping variables: I would like to drop a variable when a certain condition is met (I'm using loops to check whether a variable is not empty for each year of observation). AFAIK, - drop- and -keep- only refer to observations if being used conditionally, e.g. I cannot use "drop THISVAR if mean_THISVAR<1". So, how can I drop a variable when a certain condition is met? -------------------------------------------------------------------------------- You can easily do both. The do-file below illustrates dropping a variable (for each year of observations) has any missing values. And later it drops it if its mean is greatern than one. Joseph Coveney clear * set seed `=date("2008-07-21", "YMD")' set obs 200 forvalues i = 1/10 { generate double year`i' = uniform() if mod(`i', 2) replace year`i' = . if uniform() > 0.99 } replace year10 = 5 * year10 * * Begin here * foreach var of varlist year* { capture assert !missing(`var') if _rc { display in smcl as text "variable {result}`var' " _continue display in smcl as text "contains missing values" drop `var' continue } summarize `var', meanonly if r(mean) > 1 { display in smcl as text "variable {result}`var'" _continue display in smcl as text "'s mean is greater than one" drop `var' } } exit * * 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/* * 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/

* * 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/

**References**:**Re: st: drop variable on condition***From:*"Joseph Coveney" <jcoveney@bigplanet.com>

**Re: st: drop variable on condition***From:*Andrea Bennett <mac.stata@gmail.com>

- Prev by Date:
**Re: st: is this a missing value issue? (-if- command used)** - Next by Date:
**st: Computing GDP growth using a log-linear regression** - Previous by thread:
**Re: st: drop variable on condition** - Next by thread:
**st: TOBIT AND PWEIGHTS IN STATA 10** - Index(es):

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