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

  • Test for subpopulation vs total population

    Hi Statalist,

    Currently i'am working on my master thesis that contains the topic the 'idiosyncratic volatility puzzle'. I created 5 portfolios based on the idioysncratic volatility of the previous month. I want to look if the mean returns of these portfolios during the financial crisis, significantly differ from the total sample period mean returns. So simply said, i want to tests if the portfolio means of 2008-2009 are signifcantly different from my total sample period (2003-2020). The problem is that i have no idea which test i can use here. I (or at least think) have no independent samples, so i think a normal t-test is not applicable here?. I saw several t-tests pass by ( e.g. welch t-test, wilcoxon tests) but none of them seem to fit my problem. Is there anyone on this forum that had a comparable problem and can help me which test i need to use?. I'am new to this forum so if i made any mistakes in this post, please let me know!.
    Thanks in advance for the help!
    Marijn Wolkers