Added: 3 years ago
From: bionicturtledotcom
Views: 22,702
Sort by time | Sort by thread (beta)

Link to this comment:

Share to:

All Comments (24)

Sign In or Sign Up now to post a comment!
  • You are a brilliant teacher!

  • 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!?!

  • 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.

  • 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.

  • how did u draw the curves and what are the parameters u used for both the curves...???

  • great video, thanks

  • What synthetic voice is BionicTurtle's?

  • 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 Check out burkeyacademy com and look for videos Inference I and Inference J under Econometrics 6. This might be what you're looking for.

  • thankyou for the upload :) very clear and it helps :) xx

  • 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.

  • Thank you very much for this helpful tutorial.

  • Excellent - very clear and avoids the usual poor/simplistic explanations

  • @theoldorganplayer thank you, really appreciate that....

  • How do you create the F-distribution chart

  • Comment removed

  • The best compilation of videos I have found for learning statistical analysis for social science.

  • 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.

  • Hi EzE3000,

    Excel has a function for F distribution: = FDIST(). Hope that helps!

  • 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

  • Agree guys lectures don't explain a s..t. This guy is explaining perfectly.

  • Really great series ! I finally understood what all those t, f distributions and hypothesis testing really are for, after all these years dreading stats.

  • Great vid. Why go to lectures when theres youtube lol

  • agree... don't even know what my lecturer is talking about!!

Loading...
Alert icon
0 / 00Unsaved Playlist Return to active list
    1. Your queue is empty. Add videos to your queue using this button:
      or sign in to load a different list.
    Loading...Loading...Saving...
    • Clear all videos from this list
    • Learn more