Random Effects estimators as fGLS

Поділитися
Вставка
  • Опубліковано 25 січ 2025

КОМЕНТАРІ • 14

  • @nguyendattan638
    @nguyendattan638 Місяць тому

    thanks alot, Mr.Ben Lambert you are my hero

  • @RD-dd4cv
    @RD-dd4cv 6 років тому +2

    You are the man, i think i'm gonna pass my test in Panel Data because of your videos, keep up the good work

  • @rareclassicmovies6429
    @rareclassicmovies6429 10 років тому

    You previously mentioned that we use RE because POLS is not applicable but FE is more difficult, but here we do FE as the first step of using RE! It means that using RE is more complicated that using FE!

  • @fatmaosama4213
    @fatmaosama4213 2 місяці тому

    great explanation, thanks

  • @TÔMTIÊNYÊN
    @TÔMTIÊNYÊN Рік тому

    Thank you Sir, how can we estimate the lamda hat ?

  • @lastua8562
    @lastua8562 4 роки тому +1

    Does sigma u refer to variance in i or t or both?

  • @lastua8562
    @lastua8562 4 роки тому

    You say we estimate lambda - Should we not have a confidence interval around it?

  • @shilpidas4421
    @shilpidas4421 4 роки тому

    Can you give me the related reference book for this topic?

  • @ecozi
    @ecozi 8 років тому +1

    I don't understand what random sample in CS means. Aren't we following the same people through time when working with panel data of individuals? Then this assumption wouldn't hold, right? Can anyone explain please?

    • @lastua8562
      @lastua8562 4 роки тому

      not in pooled OLS. Here, we just have repeated cross-sections. --- well, that is what pooled OLS is. But if we have panel data, I guess we just "treat it" as it were repeated cross sections. We pool the data together as if there were no country/area/region/alpha-term included.

  • @TheMinstrel55
    @TheMinstrel55 7 років тому

    please organise these videos,they are good.

    • @lastua8562
      @lastua8562 4 роки тому

      his profile has playlists nowadays.

  • @jaminfree
    @jaminfree 9 років тому

    sigma u, not sigma mu