At 4:33 You say, "...my estimator Beta tilda is consistent..." I think you meant to say, "...my estimator gets closer and closer to the unbiased population parameter..." as it was already consistent to begin with. What is increasing here is unbiasedness rather than consistency, which is just getting reinforced. The drifting of mean towards the population mean, indicates increase in unbiasedness with an increase in sample size because of a good model, which was at least consistent even with a small sample. PS: I learned this from your videos!
How can we say that the estimator is biased or not , I can just pick up the the data of estimator with a sample size of 1 million directly and say its pretty darn unbiased , or does the sample has to be small ? and that case what should be the sample size as a percentage of the population when we have to decide if they estimator is biased or unbiased ?
Hi, There is a video on the linearity assumption if you go to my channel homepage. I would have posted the video link here, but for some reason youtube doesn't allow me to. Thanks, Ben
if we were to choose an estimator out of 2, wherein one was unbiased only and not consistent and the other one is biased and consistent, which one would we choose?
Hey Ben, has an estimator, which is biased but consistent, to be asymptotic unbiased or can there be any exceptions (which i have to know in the undergaduated course)? Thanks a lot!
Hi Ben, Great videos, they have been a serious help over the last year. I just have a question on this video: in the second scenario (biased & consistent), is it true to say the estimator becomes less biased as n tends to infinity because the estimator is consistent? Many Thanks, Keith
Hi, thanks for your comment. Yes, this is a necessary condition for beta tilde to be consistent. It is not sufficient however, since we also require that the estimator is asymptotically unbiased. Best, Ben
Ben, is the following statement true: "An estimator is consistent if and only if there exists convergence in probability", because I usually see one-way implication: if the estimator converges in probabiltiy THEN the estimator is consistent.
![I.N "HALLUCINATION" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/n5B5q1Hwt_U/mqdefault.jpg)
BEST econometrics videos!
lmao I'm studying for theoretical statistics
Hi Ben, I could not thank you enough for this and for the video series. Thanksssssssssssssss
Thank you, Ben! Greetings from Brazil!
Thanks for this video! It was very educational. Your drawings, however, will haunt me in my dreams T.T
At 4:33
You say, "...my estimator Beta tilda is consistent..."
I think you meant to say, "...my estimator gets closer and closer to the unbiased population parameter..." as it was already consistent to begin with. What is increasing here is unbiasedness rather than consistency, which is just getting reinforced. The drifting of mean towards the population mean, indicates increase in unbiasedness with an increase in sample size because of a good model, which was at least consistent even with a small sample.
PS: I learned this from your videos!
Is it possible to have an estimator which is unbiased but inconsistent? Anyone know of an example?
You are a life saver dude...
thank you so much, Ben! Great explanation
Thanks for your explanations! Now I understand much better :)
How can we say that the estimator is biased or not , I can just pick up the the data of estimator with a sample size of 1 million directly and say its pretty darn unbiased , or does the sample has to be small ? and that case what should be the sample size as a percentage of the population when we have to decide if they estimator is biased or unbiased ?
Hi, There is a video on the linearity assumption if you go to my channel homepage. I would have posted the video link here, but for some reason youtube doesn't allow me to. Thanks, Ben
Ben Lambert can you tell about the unbias and consistency of an intercept please
Can we have an estimator which is unbiased and inconsistent ?
Thank you for videos sir
Seems like the videos are not in order, is there a place I could find the correct order?
if we were to choose an estimator out of 2, wherein one was unbiased only and not consistent and the other one is biased and consistent, which one would we choose?
nicely explained! Thank you!
Can we have an estimator which is Unbiased but inconsistent ??
Thank your videos sir
Thank you !
Hey Ben,
has an estimator, which is biased but consistent, to be asymptotic unbiased or can there be any exceptions (which i have to know in the undergaduated course)?
Thanks a lot!
quite informative videos----good job
Hi Ben,
Great videos, they have been a serious help over the last year.
I just have a question on this video: in the second scenario (biased & consistent), is it true to say the estimator becomes less biased as n tends to infinity because the estimator is consistent?
Many Thanks, Keith
That is correct.
bravo!
but i have doubt in other exercise can you help me...i really appreciate it...
simply amazing
As you let the sample size tend towards infinity, does the variance of Beta_tilda go to zero, if Beta_tilda is consistent?
Hi, thanks for your comment. Yes, this is a necessary condition for beta tilde to be consistent. It is not sufficient however, since we also require that the estimator is asymptotically unbiased. Best, Ben
any chance you could do a video deriving the inconsistency in ols ??
Is it possible to have b(population)>b(tilda) so when you increase sample size the distribution function "shifts" to the right?
Question, is this series all based on Point estimates right??? nothing to do with intervals
Ben, is the following statement true: "An estimator is consistent if and only if there exists convergence in probability", because I usually see one-way implication: if the estimator converges in probabiltiy THEN the estimator is consistent.
Hi Ben. Do you do 1 to 1 tutor sessions? I am in 2nd year of Uni studying Econometrics and really struggling!
Very helpful
my guy saving my degree
Let's home mine too
anyone still here in 2020?
Hi mate
anyone still here in 2024?