Timeline for Can principal component analysis be applied to datasets containing a mix of continuous and categorical variables?
Current License: CC BY-SA 4.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| S Nov 19, 2020 at 9:42 | history | edited | chl | CC BY-SA 4.0 |
Name of the function from FactoMineR package is FAMD not AFDM
|
| S Nov 19, 2020 at 9:42 | history | suggested | UseR10085 | CC BY-SA 4.0 |
Name of the function from FactoMineR package is FADM not AFDM
|
| Nov 19, 2020 at 9:25 | review | Suggested edits | |||
| S Nov 19, 2020 at 9:42 | |||||
| S Aug 15, 2018 at 16:32 | history | suggested | Alex Firsov | CC BY-SA 4.0 |
Updated dead link to detailed description and specific implementation referenced in answer description (content moved to new location on same site; old link leads to 404 error).
|
| Aug 15, 2018 at 16:07 | review | Suggested edits | |||
| S Aug 15, 2018 at 16:32 | |||||
| Jul 17, 2014 at 16:46 | comment | added | Zhubarb |
Regarding: Although a PCA applied on binary data would yield results comparable to those obtained from a Multiple Correspondence Analysis, can we not convert a nominal categorical variable (let's say with N cardinality) into a collection of (N-1) dummy binaries and then perform PCA on this data? ( I understand there are more appropriate techniques)
|
|
| May 2, 2014 at 23:08 | comment | added | casandra | chl, thanks for the pointer to FADM. I was wondering though: once I apply FADM to a data set (obj <- FADM(x)), I can access the transformed data set easily via: obj\$ind\$coord. However, if I want to apply the same transformation to another data set, how can I do so? (This is necessary for example, if I have a train set, and I find the "principal components" from this train set, and then want to look at the test set through those "principal components"). The documentation isn't really clear on this, and the paper the function is based on is in french. | |
| Dec 28, 2010 at 7:09 | history | answered | chl | CC BY-SA 2.5 |