## PCA (Principal component analysis) in pharma

Solved
Super Contributor
Posts: 301

# PCA (Principal component analysis) in pharma

Hi guys,

I would like to know more about the applicatibility of PCA (Principal Component Analysis) in the Pharma industry.

I have experience using it in biophysics using MATLAB and I would like to start understanding the applicability of

using this method in SAS.

I know that not always the two reduce-dimensions of maximal variance obtaining with PCA, give you the maximal information of the problem,

but it is a powerful method in multivariate analysis.

Some refferences to books, examples or some papers could help too.

Thanks.

V.

Accepted Solutions
Solution
‎03-29-2012 07:25 PM
SAS Super FREQ
Posts: 9,367

## PCA (Principal component analysis) in pharma

Hi, you might want to post this question in either the STAT forum or the Clinical Trial forum:

https://communities.sas.com/community/sas_statistical_procedures

https://communities.sas.com/community/sas_and_clinical_trials

I do know that PROC PRINCOMP does Principal Component Analysis and that's about all I know. Folks in the other forums may have better insights.

cynthia

All Replies
Solution
‎03-29-2012 07:25 PM
SAS Super FREQ
Posts: 9,367

## PCA (Principal component analysis) in pharma

Hi, you might want to post this question in either the STAT forum or the Clinical Trial forum:

https://communities.sas.com/community/sas_statistical_procedures

https://communities.sas.com/community/sas_and_clinical_trials

I do know that PROC PRINCOMP does Principal Component Analysis and that's about all I know. Folks in the other forums may have better insights.

cynthia

Super User
Posts: 10,771

## PCA (Principal component analysis) in pharma

As Cynthia said .STAT forum is your better place to find answer.

You should choose the eigenvetor whose eigenvalue is great than 1,

Maybe you can get two or three eigenvetor, it is dependent on your correlative matrix.

Ksharp

🔒 This topic is solved and locked.

Discussion stats
• 2 replies
• 341 views
• 3 likes
• 3 in conversation