Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • How to extract 4 specific columns from tab results

    Hi all. I am trying to extract 4 columns from the results. The code below was designed for a binary variable Gender (1 male & 2 female). However, I am trying to modify it so that I can use it for a categorical variable race ( 1 black, 2 white, 3 asian, 4 colored). May you please assist on how I can use it for a categorical variable with more than 2 groups?

    COMMAND:

    local i = 1

    foreach var of global adult_spns_66{


    svy: tab `var', col ci

    matrix prop = e(Prop)


    local rownum = 1
    local colstart = 1
    local colend = 1


    local rows = rowsof(prop)
    foreach x in "`colstart'" "`colend'"{
    foreach num of numlist 1/`rows'{
    matrix sum`num'_`x' = prop[`num',`x']
    }

    foreach num of numlist 1/`rows'{
    if `num' == 1{
    matrix sumtotal_`x' = sum`num'_`x'
    }
    else{
    matrix sumtotal_`x' = sumtotal_`x' + sum`num'_`x'
    }
    }
    }

    matrix temp1 = prop[`rownum',`colstart']
    matrix temp1 = temp1/trace(sumtotal_`colstart')
    matrix temp2 = prop[`rownum',`colend']
    matrix temp2 = temp2/trace(sumtotal_`colend')



    matrix prop`var' = (trace(temp1))



    matrix rownames prop`var' = "`var'"



    if `i' ==1{
    matrix results = prop`var'
    }
    else{
    matrix results = (results\prop`var')
    }
    local ++i
    }



    matrix list results

  • #2
    Welcome to Statalist.

    You have posted your topic in Statalist's Mata Forum, which is used for discussions of Stata's Mata language. Your question will see a much larger audience if you post it in Statalist's General Forum. The commands you are using are actually part of Stata's matrix commands described in the outout of help matrix, which are different than commands in Stata's Mata language described in the output of help mata, even though both work with matrixes. Confusing, yes.

    Also, if you have not already done so, take a look at the Statalist FAQ linked to at the top of this page for posting guidelines and suggestions.

    Comment


    • #3
      Thank you William

      Comment

      Working...
      X