How to Perform Factor Analysis in Python(Step by Step)

[@loogsnyakwardyoh4854]

the correct column name is "Unnamed: 0". df.drop(['Unnamed: 0'], axis = 1, inplace = True) works...

[@백영래-u3x]

df = drop(['unnamed:0'], axis=1, inplace = True) This one works.

[@mariamedina5970]

This is by far the best video on the internet to explain factor analysis and all the steps necessary! Great work!

[@KindsonTheTechPro]

Glad it was helpful!

[@luisady8990]

If you have trouble w/factor analyzer: 1. Open cmd and paste: pip install factor_analyzer 2. Once it is successfully installed, paste the command again and get the file path it is downloaded in (first line after the command). 3. Follow the rest of the steps in the video.

[@KindsonTheTechPro]

Thanks for the observation!

[@inusun]

Great video, helped out a lot it processing my data using factor analysis for my data science class.

[@diepnguyen8928]

Could you explain for me why I have different results while implementing Factor Analysis in Python and STATA? There are differences in both scree plot and the value of loading factors :(( Don't know which is the correct one now :( Thank you for your support!

[@tallyxh6477]

This video litterally saved me and my Master's Dissertation. Great explanations, easy to understand and follow. thank you very much!!

[@samuelceciliochucuyamamani561]

Hello! previously I was using Minitab and obtaining the factor scoring coefficients and then graphing. I don't know how to get that with python, I hope you can help me please. Saludos

[@spef7396]

Thank you very much, final year computer science but very little experience in statistics etc. this was a massive help

[@KindsonTheTechPro]

You're very welcome!

[@elhamahmadi8673]

Hi, I have one question. if we want to use FA or PCA for feature selection of all types of data(without dropping string type of data) how can we do that? Is there any alternative to apply?

[@sharp389]

do you know of algorithms to use other criteria to select factors? like cng package in R? Horn's PA, etc.?

[@KindsonTheTechPro]

I’ll try to make a video on this

[@brettgregory3622]

Any ideas on how one could go about getting goodness of fit indices? All my FAs are done in R purely because of these fit indices but I really want to completely move over to Python.

[@xiangyu9445]

where can I find your Jupyter notebook?

[@KindsonTheTechPro]

Would you like to join and 6 Weeks Intensive Data Science Course that begins this week? Let me know on [e-mail].

[@henny628]

Hi, I am trying to analyse data with multiple factors over a series of time. Is there a way I can consult you off youtube?

[@tarunsiga3610]

Could you explain me on what basis the factors were selected in FactorAnalyzer

[@KindsonTheTechPro]

Nice question but the explanation is not trivial. We consider the factor loading of each variable which tells us the entent to which the variable relates with the factor. We then select factors with highest loadings. I'll recomment you read up Principal Components Analysis(PCA) as it helps you understand FA. www.kindsonthegenius.com/pca-tutorial-1-how-to-perform-principal-components-analysis-pca/ www.kindsonthegenius.com/principal-components-analysispca-in-python-step-by-step/ www.kindsonthegenius.com/basics-of-factor-analysis-for-data-scientists/ Question, What is Factor Analysis kzbin.info/www/bejne/qWPJl557gb2mfsU

[@williamkanaan7554]

When I type fa.fit(df), it tells me there is an error "ValueError: Found array with 0 sample(s) (shape=(0, 38)) while a minimum of 1 is required by FactorAnalyzer." but when I check my document there is not ligne with 0 values, what can I do please?

[@stepheneaster3166]

Hi, just finding this great video on FA, thanks for posting it. When I use the varimax orthogonal rotation and then use transform() to 'score' the dataframe with the factor model, the resulting factors do not have 0 correlation as expected from an orthogonal transformation. I'd like to take those factors as inputs into a clustering exercise but want non correlated factors (similar to PCA). Thanks in advance for any thoughts on where I might be going awry.

[@austin_adetunji]

Wow, this is great. i wish i can like this video multiple times

[@rajeshghosh4344]

Even after appending the site package directory the factor_analyzer is not importing. Can you help?

[@Sudo007]

Please, tell me what the values in the matrix signify.

[@TieMaxx]

Answers to a questionnaire about personality (Big-Five traits). A1 through 5 represent questions about agreeableness, N is neuroticism, C is conscientiousness, E is extraversion and O is openness to experience. 1 represents an answer of -2 and 5 is actually +2. It does not really matter though, because they are all getting standardized by the standard variation and the mean. For example one question might be "I am interested in talking about abstract topics". -2 (in the matrix: 1.0) would mean that the statement does not describe you at all, while +2 (in the matrix: 5.0) means, it describes you perfectly. etc. etc.

[@aysenzumrutsonmez3699]

I had an ModuleNotFoundError for factor_analyzer in the first step. How i solve this error? Please help me

[@matangaelisha2116]

i cant see the link to the dataset please

[@chavianddavid]

Is there something called "Factor Scores" as well?

[@vineetsodhani5398]

How does one load the factor analyzer package? Where does one find the package?

[@Sudo007]

pip install factor_analyzer

[@Rskingfisher]

In 15:51-16:07 you said ev=eigenvectors and v=eigenvalues. But during the scree plot, you said ev = eigenvalues. Which is correct?

[@twinblade02]

ev is the eigenvectors.

[@Lernal]

good factor analysis for python

[@nnn-bb5tf]

Thanks for this excellent video

[@KindsonTheTechPro]

So nice of you

[@kennedybwire9435]

Hi, that is great work.

[@uniongrob8194]

Helpfull video, thanks!

[@phanthilanhuong1783]

Thank you for your sharing and fantastic guiding

[@quiosaevaristo7746]

Hi, that is an amazing video.

[@inspiritlashi9994]

This is such a great video.. Thank you so much

[@KindsonTheTechPro]

You're welcome! And do remember to subscribe 😃

[@javiercmh]

very useful as it is!!! thanks a ton!

[@sindimazibuko1399]

Can someone please help me figure out the name of the test I need to perform to see if my data is suitable for factor analysis? I can not seem to hear the name. Thank you in advance.

[@qudsiyajan6485]

KMO and Bartlett's test of sphericity can help you in knowing if data is suitable for factor analysis or not.....

[@veronicam2102]

Thanks!! this video was very helpful to me :D

[@josephkarianjahi1467]

Good work!

[@rajdeepkumarnath8944]

Thank You!

[@c.e.g.o4797]

Thank you, bro.

[@lucasmigge3159]

thank you very much

[@al702893]

You just copied datacamp's tutorial, including comments