Factor Analysis - model representation

[Josefk]

Long time ago I bought a book of Factor Analysis that tries to explain step by step this methodology. I read it during few weeks, still confused. Five minutes here. Just five minutes has been more profitable than reading that book for few weeks. Excellent. Thank you

[kausalyaakannan]

Planning to make a deal with my university management to pay half of the tuition fees to youtubers for making us understanding something better. Thanks a lot for this helpful video. I personally think lecture notes shall be replaced with tutorial videos from youtubers.

[foedeer]

This was so well explained and clarified the analysis which i understood incorrectly. Thank you!

[fangzhang]

Amazingly explained, now I just realized how bad my professor giving a lecture compared to this...

[yaweli]

Great video. Thanks for teaching people like me who can’t cope with work and school. Do you by chance have a clinical data for this example of PCA and factor analysis?

[funfair-bswf]

Hi, thank you for this very instructive video. I just have a (perhaps dumb) question. If get things right, in the equation form of y_i, eta_i is the vector representing the unobserved variable, and lambda_ij is a constant. If so, I wonder what is this constant. I get it relates to/represents the causal link between the observed and unobserved variables, but (if my question makes any sense) how do you determine lambda ? Which video of the serie should I watch to have an answer to that question ?

[zombiepandax]

Thanks a lot for this well explanation! I am just a bit confused by that you say that the first part of the equation would be communality - in my university statistics course we learned that you obtain the communality for each item by squaring the loadings and adding them up (which obviously leads to different results than if you multiply them with the factors)? Or are there different meanings for the term "communality"? but apart from that really nice job! :)