Announcement

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

  • understanding what e(pi) means in the context of stored results in the fmm package

    Hello,

    I am using FMM in stata 16, I am working with someone else's code, who used the previous fmm user generated package, in stata 14.

    The code grabs the mean and sd from the fmm estimations. It also grabs the e(pi1_est)
    I am just wondering what this e(pi_est) refers to? I cannot find it in the list of stored results in the fmm manual in stata, or in the result table.
    See code below:

    gen var_mean = _b[component2:_cons]
    gen var_sd = e(sigma2_est)

    gen var_p = e(pi1_est)

    Any help would be much appreciated.

  • #2
    Welcome to Statalist.

    From within Stata, I ran
    Code:
    search fmm
    and scrolled down to the user-written packages. I clicked on fmm and on the page that opened, I clicked on fmm.hlp to read the help for the user-written version of the command. In that we find
    Code:
    Saved results
    
        In addition to standard results saved by maximum likelihood procedures in e(), fmm saves the
        following scalars:
    
            e(parname_est) parameter estimate
            e(parname_se) standard error of estimate
    
        where parname denotes either a scale parameter or a mixing probability parameter.
    so presumably your predecessor had a parameter named pi1 and e(pi1_est) is the estimated value of that parameter.

    With that said, you have accidentally posted your topic in Statalist's Mata Forum, which is used for discussions of Stata's Mata language, which is different than Stata's command language, and different than Stata's matrix commands. Your question would have seen a more appropriate, and much larger audience if you had posted it in Statalist's General Forum.

    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
      Thanks so much William,
      I think I will just have to use the previous user package and run the code to see what happens.
      My apologies for the miss posting. I will be sure to have a look at the FAQ now.

      Comment

      Working...
      X