Announcement

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

  • ppm use for imputation

    Hi all,
    i have 57 variables and many of them have missing values. i have tried mi with chained equations but it introduces negative values. i am trying to use ppm and i am using the code below but i get errors:
    mi impute: invalid method ppm
    Method ppm is not one of the officially supported mi impute methods. If you are adding a new user-defined method to mi impute, you need to create a parser named
    mi_impute_cmd_ppm_parse.ado and an imputer named mi_impute_cmd_ppm.ado; see mi impute user


    any suggestion what am i adoing wrong

    Code:
    mi set mlong
    mi register imputed cnv bmi waistcm hipcm whratio glucosemmoll hba1c insulinmcunitml cpeptidengml homarir homab hdl ldl tcholesterol triglyceridemmoll systolicbp diastolicbp pulse  bodyweighkg heightinm albumingl alkalinephosphataseul altgptul astgotul calcium calciumcorrected chloridemmoll bicarbonatemmoll bilirubintotalumoll creatinekinaseul creatinineumoll vitD estradiopmlll ferritinmcgl folatenmoll freethyroxinepmlol freetriiodothyroninepmll glutamyltransferaseul homocysteineumoll ironumolml magnesiummmoll  phosphorusmmoll potassiummmoll shbgmnmoll sodiummmoll testosteronetotalnmoll tshmiul tibc totalproteingl ureammoll uricacidumoll vitaminb12pmoll ffmkg fat fatmasskg  fatmassindexkgm2 vatmasskgs bai  
    
    mi impute ppm cnv bmi waistcm hipcm whratio glucosemmoll hba1c insulinmcunitml cpeptidengml homarir homab hdl ldl tcholesterol triglyceridemmoll systolicbp diastolicbp pulse  bodyweighkg heightinm albumingl alkalinephosphataseul altgptul astgotul calcium calciumcorrected chloridemmoll bicarbonatemmoll bilirubintotalumoll creatinekinaseul creatinineumoll vitD estradiopmlll ferritinmcgl folatenmoll freethyroxinepmlol freetriiodothyroninepmll glutamyltransferaseul homocysteineumoll ironumolml magnesiummmoll phosphorusmmoll potassiummmoll shbgmnmoll sodiummmoll testosteronetotalnmoll tshmiul tibc totalproteingl ureammoll uricacidumoll vitaminb12pmoll ffmkg fat fatmasskg fatmassindexkgm2 vatmasskgs bai= age, add(10)
    
    note: missing-value pattern is monotone; no iteration performed
    stat 15.1 mac

    thanks in advance

  • #2
    I suspect you want to use predictive mean matching. In that case, the abbreviation is pmm rather than ppm
    ---------------------------------
    Maarten L. Buis
    University of Konstanz
    Department of history and sociology
    box 40
    78457 Konstanz
    Germany
    http://www.maartenbuis.nl
    ---------------------------------

    Comment


    • #3
      You may wish to read this thread, particularly #2.
      Best regards,

      Marcos

      Comment


      • #4
        excellent thanks.

        Comment

        Working...
        X