Sorry for my lack of understanding but, why do we want the larger variance in the numerator? How is an F distribution plotted as values from less than 1 to infin. if an F-stat given by our test is always greater or equal to 1?
How would the distribution be plotted given two underlying populations with the same variances? I assume the smaller the sample sizes n the greater the skewness of the distribution (degrees of freedom).
I struggle to understand how the distr. tends to normality. Ty.
I dont know if that makes sense. Type 1 error? By having 9df with the same 90% confidence, you can conclude significant difference?... in spite of the fact that 19df should provide you with more accurate data and does not make that conclusion.
The explanation is good, but it explains the wrong thing with this particular application. The exact F distribution applies if the underlying data is IID, but in this application it is time series. So the inference will be valid asymptotically, but not in finite samples.
This is great help! I am taking classes onlin through UoP and I wish you were teaching it. The only thing I do not understand is how you create the F distribution chart. Please help
Really great series ! I finally understood what all those t, f distributions and hypothesis testing really are for, after all these years dreading stats.
You are a brilliant teacher!
mismailjangda 2 months ago
Thanks for the vid. A bit confused about the difference between what the f-test shows and what we conclude from the t-test.
After watching your video it seems:
The f-test is to s as the t-test is to xbar!?!
Backtothe1920s 3 months ago
It seems that your H1is : varience1>varience2
but not varience 1 not equal varience 2.
if you are testing varience 1 not equal varience 2, then you should find the F value whose first parameter equals alpha/2, but not alpha.
yzy010858 3 months ago
Sorry for my lack of understanding but, why do we want the larger variance in the numerator? How is an F distribution plotted as values from less than 1 to infin. if an F-stat given by our test is always greater or equal to 1?
How would the distribution be plotted given two underlying populations with the same variances? I assume the smaller the sample sizes n the greater the skewness of the distribution (degrees of freedom).
I struggle to understand how the distr. tends to normality. Ty.
SlanderInYoFace 4 months ago
how did u draw the curves and what are the parameters u used for both the curves...???
abarashyasir1 5 months ago
great video, thanks
jamesmungall 5 months ago
What synthetic voice is BionicTurtle's?
ASSHOLELA 7 months ago
It would be really awesome if you could make a video which easily explains how the normal, chi, t and f distribution relates to each other!
HRSDKK 8 months ago
@HRSDKK Check out burkeyacademy com and look for videos Inference I and Inference J under Econometrics 6. This might be what you're looking for.
BurkeyAcademy 7 months ago
thankyou for the upload :) very clear and it helps :) xx
TimmysMummy 8 months ago
I dont know if that makes sense. Type 1 error? By having 9df with the same 90% confidence, you can conclude significant difference?... in spite of the fact that 19df should provide you with more accurate data and does not make that conclusion.
CogitoErgoCogitoSum 1 year ago
Thank you very much for this helpful tutorial.
emaghsoodlou 1 year ago
Excellent - very clear and avoids the usual poor/simplistic explanations
theoldorganplayer 1 year ago 4
@theoldorganplayer thank you, really appreciate that....
bionicturtledotcom 1 year ago
How do you create the F-distribution chart
williamrgre 1 year ago
Comment removed
oocyrpticoo 2 years ago
The best compilation of videos I have found for learning statistical analysis for social science.
dmglanz 2 years ago
The explanation is good, but it explains the wrong thing with this particular application. The exact F distribution applies if the underlying data is IID, but in this application it is time series. So the inference will be valid asymptotically, but not in finite samples.
FancyStasik 2 years ago
Hi EzE3000,
Excel has a function for F distribution: = FDIST(). Hope that helps!
bionicturtledotcom 3 years ago
This is great help! I am taking classes onlin through UoP and I wish you were teaching it. The only thing I do not understand is how you create the F distribution chart. Please help
EzE3000 3 years ago
This has been flagged as spam show
this is good, theres some other good videos on tutormetvdotcom. It helped me with my midterms
ljuarez714 2 years ago
Agree guys lectures don't explain a s..t. This guy is explaining perfectly.
machvis 3 years ago
Really great series ! I finally understood what all those t, f distributions and hypothesis testing really are for, after all these years dreading stats.
a3306338 3 years ago
Great vid. Why go to lectures when theres youtube lol
Dimenchemist 3 years ago
agree... don't even know what my lecturer is talking about!!
pompomjangkrik 3 years ago