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

  • adding summary weight in metabolic score

    I am trying to generate a metabolic syndrome score for the children aged 11 to 12 years using PCA. I run the PCA with the standardised z-score of all five metabolic components But I confused on one statement mentioned in the article that I am following "Our PCA identified
    2 principal components that were summed, with weights determined by the relative amount of variance, explained to generate a total MetSrisk score".
    I am also missing Stata command for final analysis " we internally standardise the final met score to get mean 0 and SD 1 so that regression coefficient can be explained by SMD

    I USED THE FOLLOWING COMMAND. It will be highly ap[appreciated if you can help me out with missing command lines. I am badly stuck at this point.

    global xlist Z2TotChol Z2Glucose Z2HDL Z2TG fcbmizc
    global ncomp 2

    ** PCA
    pca $xlist
    screeplot, yline (1)

    ** PCA Component
    pca $xlist, comp($ncomp)
    pca $xlist, comp($ncomp) blanks(0.3)

    rotate, varimax
    rotate, varimax blanks(0.3)

    rotate, promax
    rotate, promax blanks(0.3)

    estat loadings
    estat residuals, fitted
    estat loadings, cnorm(eigen)

    predict Pc1 Pc2, score
    gen PCAScore=Pc1 + Pc2

    Thanks in advance

  • #2
    Couldn't you write the authors and ask them to share their code?


    • #3
      I wrote but they did not reply! Very unfortunate


      • #4
        Indeed. Try contacting the editor or publisher of the journal where the paper appeared. Maybe the authors had to deposit the datasets and code as a condition of publication.


        • #5
          Originally posted by Tahmina Begum View Post
          I wrote but they did not reply! Very unfortunate
          I was wondering if there is any update on this? Thanks!


          • #6
            Yes, I one author confirmed me that they used proportion of each PCs as weight