Visualization of PCA shape change

[theropod1]

Hi!
Love your package!

I have a question/perhaps a feature request:

Namely, I’m working on visualization of PCA morphospaces using the PCcontrib() function.

For example:
PCcontrib(my_pca_object, nax=1:1, sd.r=c(-1))

I understand this will return me the shape at -1 standard deviation on PC1, and at 0 for all the other PCs.

What I am wondering is if there is a way to also visualize shapes that are at some arbitrary point in shape space, e.g. at -1 on PC1 and -1 on PC2.
Package Geomorph does something similar with its shape.predictor(), which allows to predict the shape for any combination of predictor variables (i.e. PC values).
Is there a way to accomplish something like this in Momocs?

Many thanks!

[vbonhomme]

Ideally yes. So far no. But soon, say this autumn, yes ! Le dim. 10 juil. 2022 à 19:40, theropod1 ***@***.***> a écrit :

…

Hi! Love your package! I have a question/perhaps a feature request: Namely, I’m working on visualization of PCA morphospaces using the PCcontrib() function. For example: PCcontrib(my_pca_object, nax=1:1, sd.r=c(-1)) I understand this will return me the shape at -1 standard deviation on PC1, and at 0 for all the other PCs. What I am wondering is if there is a way to also visualize shapes that are at some arbitrary point in shape space, e.g. at -1 on PC1 and -1 on PC2. Package Geomorph does something similar with its shape.predictor(), which allows to predict the shape for any combination of predictor variables (i.e. PC values). Is there a way to accomplish something like this in Momocs? Additionally Many thanks! — Reply to this email directly, view it on GitHub <#224>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABECQFQSRGS734HLXH44SR3VTMDJ7ANCNFSM53FKTXTQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>