You all prolly dont give a shit but does anybody know a method to get back into an instagram account? I stupidly forgot the password. I love any assistance you can offer me.
@Enzo Andre Thanks so much for your reply. I found the site thru google and im in the hacking process atm. I see it takes quite some time so I will get back to you later with my results.
FYI - per my notes, there are two ways of removing fixed effects - one method he's talking about sounds like "demeaning" and another method is first differencing. Correct me if I'm wrong :)
First differencing is only equivalent to the demeaning approach to removing fixed effects if there are exactly two time periods. Otherwise, demeaning is generally preferred to first differencing (in many but not all cases) because it makes weaker assumptions about serial correlation in the errors.
Hi Nick. I have just started following your causality series, and it really is wonderful. I just wonder, in the case of fixed effect, does it could unintentionally control the collider and thus make a bias? let's say for the height vs basket ability in the NBA example (assuming there is height variation in each year, while there is no variation in NBA status across years)
Thank you! It would be an unusual case where fixed effects introduce collider bias, since for that to be the case, one of those fixed-over-time characteristics would have to be caused by two separate variable-over-time characteristics. It's certainly possible that there is a collider bias problem in the analysis anyway that the fixed effects don't solve, though. In the NBA example, there's already a collider bias problem having ot do with the ability to get into the NBA, and fixed effects would not resolve the issue.
Hi Nick I am doing some regressions and have models where i include year FE and others where i inculde year FE and country FE. Adding country fixed effects removes all statistical significance from my models. Would it then be appropriate to say: "Interestingly the inclusion of country fixed effects removes all significance from all the models. This means that by controlling for country specific unobservable in the model there is no longer any significant results. This could mean that there is a variable that explains how voting coincidence changes over time that is not included in our models. This could be a variable that isn’t included in our dataset or even a variable that is not able to be measured." Thanks.
Close. If it's the country fixed effects that do it, it would be more about a country effect that IS constant over time. Alternatively, make sure you have enough within variation in whatever you're studying to be able to actually use FE. If your treatment is almost entirely constant over time within country, then your results will be highly noisy with country FE, and could be insignificant for that reason.
Is age considered something fixed about a person? I know age change over time, but it changes for all individuals in the same way. I can't figure out if I should control for age in my two-way fixed effects model?
It isn't fixed because it changes over time. Birth year, on the other hand, is fixed and would be accounted for by individual fixed effects. And age is just the difference between current year and birth year. So in a two-way fixed effects model, the combination of individual fixed effects with time fixed effects should control for age.
Thanks a lot for the video. Let‘s suppose I have a cross-sectional dataset with each observation being a different firm. The firm’s can be from two different industries. Did I understand you correctly that it is impossible to make a firm fixed effects regression on this? And also no industry fixed effects regression as we only have two industries? Would appreciate a response from you a lot! Many thanks!
You couldn't do firm fixed effects, you need multiple observations per firm to do firm fixed effects. You could do industry fixed effects though (although it's more just a regular ol control for industry)
@@NickHuntingtonKlein Hey Nick thanks so much for the fast reply. Thanks to your videos I am finally beginning to start understanding the topic of fixed effects! Just a quick follow up question regarding the industry: A simple dummy variable for the industry as a control variable in the regression would be enough, right?
Your videos are incredibly helpful!
Always very helpful, huge fan if yours! Cheers from Barcelona
Thank you very much, form Kiribati islands
very interactive and engaging, thank you very much
Hi Nick, i'd like thank you for this serie of videos. They are both supportive and helpful to people like me - someone new in causality.
Thank you very much! Glad you like them
You all prolly dont give a shit but does anybody know a method to get back into an instagram account?
I stupidly forgot the password. I love any assistance you can offer me.
@Kaiser Zander instablaster =)
@Enzo Andre Thanks so much for your reply. I found the site thru google and im in the hacking process atm.
I see it takes quite some time so I will get back to you later with my results.
@Enzo Andre It did the trick and I actually got access to my account again. I am so happy!
Thank you so much, you really help me out!
You're trully helpful! Thanks a lot
FYI - per my notes, there are two ways of removing fixed effects - one method he's talking about sounds like "demeaning" and another method is first differencing. Correct me if I'm wrong :)
First differencing is only equivalent to the demeaning approach to removing fixed effects if there are exactly two time periods. Otherwise, demeaning is generally preferred to first differencing (in many but not all cases) because it makes weaker assumptions about serial correlation in the errors.
@@NickHuntingtonKlein Thank you!
nice explanation
Hi Nick. I have just started following your causality series, and it really is wonderful. I just wonder, in the case of fixed effect, does it could unintentionally control the collider and thus make a bias? let's say for the height vs basket ability in the NBA example (assuming there is height variation in each year, while there is no variation in NBA status across years)
Thank you!
It would be an unusual case where fixed effects introduce collider bias, since for that to be the case, one of those fixed-over-time characteristics would have to be caused by two separate variable-over-time characteristics.
It's certainly possible that there is a collider bias problem in the analysis anyway that the fixed effects don't solve, though. In the NBA example, there's already a collider bias problem having ot do with the ability to get into the NBA, and fixed effects would not resolve the issue.
@@NickHuntingtonKlein Thank you for your enlightening answer Nick!
Great video! Thank you for the explanation! :D I will check out the other videos on your channel!
Thank you so much. Great explanation.
Thanks. Really good explanation!
Hi Nick
I am doing some regressions and have models where i include year FE and others where i inculde year FE and country FE. Adding country fixed effects removes all statistical significance from my models. Would it then be appropriate to say:
"Interestingly the inclusion of country fixed effects removes all significance from all the models. This means that by controlling for country specific unobservable in the model there is no longer any significant results. This could mean that there is a variable that explains how voting coincidence changes over time that is not included in our models. This could be a variable that isn’t included in our dataset or even a variable that is not able to be measured."
Thanks.
Close. If it's the country fixed effects that do it, it would be more about a country effect that IS constant over time. Alternatively, make sure you have enough within variation in whatever you're studying to be able to actually use FE. If your treatment is almost entirely constant over time within country, then your results will be highly noisy with country FE, and could be insignificant for that reason.
Is age considered something fixed about a person? I know age change over time, but it changes for all individuals in the same way. I can't figure out if I should control for age in my two-way fixed effects model?
It isn't fixed because it changes over time. Birth year, on the other hand, is fixed and would be accounted for by individual fixed effects. And age is just the difference between current year and birth year. So in a two-way fixed effects model, the combination of individual fixed effects with time fixed effects should control for age.
Thanks a lot for the video. Let‘s suppose I have a cross-sectional dataset with each observation being a different firm. The firm’s can be from two different industries. Did I understand you correctly that it is impossible to make a firm fixed effects regression on this? And also no industry fixed effects regression as we only have two industries? Would appreciate a response from you a lot! Many thanks!
You couldn't do firm fixed effects, you need multiple observations per firm to do firm fixed effects. You could do industry fixed effects though (although it's more just a regular ol control for industry)
@@NickHuntingtonKlein Hey Nick thanks so much for the fast reply. Thanks to your videos I am finally beginning to start understanding the topic of fixed effects! Just a quick follow up question regarding the industry: A simple dummy variable for the industry as a control variable in the regression would be enough, right?
@@ramonkonig5166 correct. And thanks!
GG!!!!! Thanks
Very useful. Thanks :)
lifesaver
would you be willing to share the animation code?
All the code for these slides, including the animation, is at the GitHub repository : github.com/NickCH-K/EconometricsSlides