Fixed Effects, First Differences and Pooled OLS - intuition

แชร์
ฝัง
  • เผยแพร่เมื่อ 23 ม.ค. 2025

ความคิดเห็น • 50

  • @saralaabid6118
    @saralaabid6118 5 ปีที่แล้ว +54

    Ben, I know you have been told it a million times, but I can't emphasise enough how grateful I am for these very didactic videos! God bless

    • @lastua8562
      @lastua8562 4 ปีที่แล้ว +2

      Is it not usually the case that an increase in mathematic knowledge is associated with a decrease in beliefs in God?

    • @GeorificaArty
      @GeorificaArty 4 ปีที่แล้ว

      @@lastua8562 unless u believed 1=3 no i think ur fine

    • @oliverxavier1493
      @oliverxavier1493 3 ปีที่แล้ว

      I guess im asking the wrong place but does anyone know of a trick to get back into an instagram account..?
      I stupidly lost my password. I would appreciate any tips you can give me.

    • @kyriekody1946
      @kyriekody1946 3 ปีที่แล้ว

      @Oliver Xavier instablaster ;)

    • @oliverxavier1493
      @oliverxavier1493 3 ปีที่แล้ว

      @Kyrie Kody Thanks so much for your reply. I got to the site through google and Im in the hacking process atm.
      Takes quite some time so I will get back to you later with my results.

  • @kunstkt
    @kunstkt 11 ปีที่แล้ว +4

    I was so caught up watching your "full course" that I didn't realize you were now uploading stuff for panel data. Once again, you'll be my best friend this semester. Big thanks!

    • @SpartacanUsuals
      @SpartacanUsuals  11 ปีที่แล้ว

      Hi, many thanks for your comment, and kind words. I am glad to hear that the videos are useful! If you have any ideas for further topics etc. that you would like covered, then please let me know. Best, Ben

    • @kunstkt
      @kunstkt 11 ปีที่แล้ว

      Ben Lambert
      I'm already looking at a couple of online (text) tutorials on time series decomposition, but it would be great to hear your approach on that subject.

    • @SpartacanUsuals
      @SpartacanUsuals  11 ปีที่แล้ว

      Hi, thanks for your message. If you could be more specific about time series decomposition that would be great - I can then add this topic to my list of things I intend to cover. Again, thanks for your input here! Best, Ben

    • @kunstkt
      @kunstkt 11 ปีที่แล้ว

      Ben Lambert
      I mean extracting trends and cycles from series. Thanks Ben.

  • @wanjadouglas3058
    @wanjadouglas3058 3 ปีที่แล้ว

    Well done video Ben. The use of an example makes a whole difference.

  • @joyyomitokosi
    @joyyomitokosi 7 ปีที่แล้ว

    ben you do not even realised how much you saved me ! thank you

  • @Carterv3
    @Carterv3 10 ปีที่แล้ว +10

    Great video Ben, everything that I wanted answered about pooled OLS and fixed effects was answered. Really great.

    • @SpartacanUsuals
      @SpartacanUsuals  10 ปีที่แล้ว +2

      Hi, glad to hear it helped! Thanks, Ben

  • @sideman4501
    @sideman4501 8 ปีที่แล้ว

    Hi Ben, wanted to thank you for your videos and especially this simple diagram showing intuition behind how pooled ols and FE give different correlations. Helped hugely with dissertation ! Owe a lot to your videos. Thanks.

  • @PaulJohn204
    @PaulJohn204 11 ปีที่แล้ว +1

    What an excellent video, just what I was looking for. A simple and quick, but still highly comprehensible overview of FE vs OLS. It was also fairly topical considering I'm from Brighton, thanks! 

    • @SpartacanUsuals
      @SpartacanUsuals  11 ปีที่แล้ว

      Hi, thanks very much for your message, and kind words. I am actually from (originally) very near Brighton in Sussex! All the best with the studies, Ben

  • @Wakanda_Wang
    @Wakanda_Wang 11 หลายเดือนก่อน

    Explanation is very crystal clear! It very helpful , thanks a lot!

  • @saschaadam3328
    @saschaadam3328 10 ปีที่แล้ว +2

    Hey Ben, thanks for the cool video! I was able to understand the topic within a few minutes instead of reading econometrics books for hours :-)

  • @Lukas-md3bk
    @Lukas-md3bk 9 ปีที่แล้ว +1

    So well explained! Perfectly for econometrics at university! awesome!

  • @datajunkie3427
    @datajunkie3427 11 ปีที่แล้ว +1

    This helps me a lot in understanding a paper I am reading. Thanks!

    • @SpartacanUsuals
      @SpartacanUsuals  11 ปีที่แล้ว +1

      Hi, glad to hear it helps! Many thanks for your comment. Best, Ben

  • @binkshimself
    @binkshimself 5 ปีที่แล้ว

    Really great video. Simple and intuitive.

  • @ericzhang4732
    @ericzhang4732 4 ปีที่แล้ว

    For god sake(with British accent), you really saved my ass !!!!!!! Thank you very much !!!!!!!!

  • @Mattmfu10
    @Mattmfu10 ปีที่แล้ว

    Brilliant! and I don't mean it in the way the Brits do (cool) - I mean it in the way Americans do(very smart). I have been studying FE/FD models for a few weeks and I have not seen a more intuitive snapshot than this.

  • @sunnyboy288
    @sunnyboy288 6 ปีที่แล้ว

    Can you please clarify what you say at 6:04? Line through the center of average of all the observations or just the Bristol city's average? "And the line should pass close to other two observations" . Should it pass closest to two observations of only Bristol city? or the two observations of all the cities?

  • @andreassihotang346
    @andreassihotang346 4 ปีที่แล้ว

    Hi Ben, I just started watching your lectures on TH-cam since last years, and I really enjoy watching all of it. It's been really helpful for me to improve my understanding on econometrics and related topics, as I am doing my graduate study. Related to this video, is it possible to show how the line in the graph looks like if we estimate the effect of unemployment on crime rate with Random Effects? Thanks!

  • @SomethingSoOriginal
    @SomethingSoOriginal 9 ปีที่แล้ว

    Really helpful Ben, intuition videos are great!

  • @TYDB66
    @TYDB66 4 ปีที่แล้ว

    Essentially, this is very helpful

  • @황인영-t9c
    @황인영-t9c 7 ปีที่แล้ว

    Thank you! This video is very intuitive !!

  • @nikkitagupta319
    @nikkitagupta319 ปีที่แล้ว

    Thank you so much Sir for these lecture series! Explained so well, I like and watch your lectures a lot!
    Sir, is there any theory- based assignment also to practice questions based on panel data (as is available in case of time series). If yes, then kindly share the link🙏

  • @LisaHoving
    @LisaHoving 9 ปีที่แล้ว

    You always explain so very well =) thank you!

  • @boldmaabat-erdene3866
    @boldmaabat-erdene3866 4 ปีที่แล้ว

    Thank you, Ben!

  • @edddc
    @edddc 11 ปีที่แล้ว +1

    Lovely stuff, thanks Ben.

    • @SpartacanUsuals
      @SpartacanUsuals  10 ปีที่แล้ว

      Hi, glad to hear it was helpful. All the best, Ben

  • @daphneashba
    @daphneashba 5 ปีที่แล้ว

    So first differences isn’t like a derivative wipes the constant or fixed effects.The constant term is gone but you are saying that take time unchanging similar properties for all of three cities?

  • @pierosraffa8208
    @pierosraffa8208 5 ปีที่แล้ว

    Hi Ben. What may be the reason of this message after Hausman test (HAUSMAN TEST SET TO ZERO)

  • @martineuwe4144
    @martineuwe4144 8 ปีที่แล้ว

    Very helpfull, thanks! But I still have a question.. So in this example what is the difference between year fixed effects and say city fixed effect? Or is this just terminology?

  • @lorenzorimella468
    @lorenzorimella468 7 ปีที่แล้ว

    Thank you very much! So helpful!

  • @AlexJones-os8nb
    @AlexJones-os8nb 9 ปีที่แล้ว +1

    Thanks for the great video! Would this be called city fixed effects, or time fixed effects?

  • @danielrichards5889
    @danielrichards5889 10 ปีที่แล้ว

    Hi, Great Vids, If you had the results from a panel model: RENTt = 1.7 + 2.6GDPt - 1.3SIR t-2
    How would you go about answering how to calculate fixed affects in the above model?
    Thanks

  • @danielnobregadelimarego9914
    @danielnobregadelimarego9914 9 หลายเดือนก่อน

    what a legend

  • @yunusemrebilgili5847
    @yunusemrebilgili5847 2 ปีที่แล้ว

    well, that was gooodddd

  • @cach_dies
    @cach_dies 7 ปีที่แล้ว +3

    Please teach at my university

  • @bobcrunch
    @bobcrunch 10 ปีที่แล้ว

    Your thesis is that the crime rate only depends on the unemployment rate. There could be other factors. Maybe the population became more religious which has much more influence to decrease the crime rate.

  • @danaronowicz635
    @danaronowicz635 3 ปีที่แล้ว +1

    When you spend 50k on tuition fees for a "good" university and you end up learning what you need because you're university didn't teach it to you for free on youtube. Such jokes, and tks Benny

  • @StayForTheFireworks
    @StayForTheFireworks 7 ปีที่แล้ว

    You're saving my ass!!!