Timeline for Why is covariance matrix not positive-definite when number of observations is less than number of dimensions?
Current License: CC BY-SA 3.0
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 28, 2018 at 13:42 | vote | accept | pierre | ||
| Sep 6, 2017 at 14:10 | answer | added | whuber♦ | timeline score: 4 | |
| Mar 4, 2016 at 5:39 | comment | added | Gottfried Helms | I wouldn't know why you couldn't at least look via principal components and rotations on your data to see patterns or to reduce to a smaller number of components? (Of course a true factor analytic model expects itemspecific variance which you cannot model with your data) | |
| Mar 3, 2016 at 0:06 | comment | added | amoeba | +1 but your second question (about the Matlab code) is off-topic here. | |
| Mar 3, 2016 at 0:04 | history | edited | amoeba | CC BY-SA 3.0 |
light editing
|
| Mar 2, 2016 at 10:34 | comment | added | pierre | Then in my case which method of dimension reduction would you suggest? And can you also propose a standard matlab code for that method? | |
| Feb 26, 2016 at 11:56 | comment | added | ttnphns |
This is a theoretical problem (see Pt 6). Due to relatively low n correlations cannot enough differentiate from one another and do not allow the factor model to play in full accordingly. So forget FA. It is good to have n>p at least 3-5 times, practically.
|
|
| Feb 26, 2016 at 11:49 | comment | added | pierre | @ttnphns; Is there any solution to the problem or simply I have to forget factor analysis? | |
| Feb 25, 2016 at 19:19 | comment | added | ttnphns | You cannot do factor analysis (most algorithms and implementations won't allow) on a singular correlation matrix (and when n<p, it is but singular) as well as negative-definite matrix (which could appear sometimes with pairwise deletion of missng values). | |
| Feb 25, 2016 at 12:21 | history | edited | Andy | CC BY-SA 3.0 |
deleted 1 character in body
|
| Feb 25, 2016 at 11:37 | review | First posts | |||
| Feb 25, 2016 at 12:21 | |||||
| Feb 25, 2016 at 11:36 | history | asked | pierre | CC BY-SA 3.0 |