Natural feeling domain-specific language for building structural equation models in R for estimation by covariance-based methods (like LISREL/Lavaan) or partial least squares (like SmartPLS)
58
stars
19
forks
source link
Error: 'subscript endogenous_items must be a simple vector, not a matrix' #331
I have thoroughly enjoyed using your hugely functional package for finishing up my thesis, however I have recently revised my model and have been unable to run the predict_pls function - this is quite important for my analyses.
The error I'm getting is, 'Error in ordered_data[trainIndexes, endogenous_items]:
Can't subset columns with endogenous_items.
x Subscript endogenous_items must be a simple vector, not a matrix'.
I have tried cloning the pls_predict code and executing this, which worked before. However I haven't had much luck this time round. Any idea where or what the endogeneity is, or if I can change the matrix to a simple vector as it says?
I was wondering if you could highlight where I could amend this so I can finish my analyses? Just to note my seminr model is a mediated model - I also have an unmediated model, and have tried this too, but the error is different:
'error in sample.int(length(x), size, replace, prob):
invalid 'size' argument.
Hope you can help with this, have had so much fun and success with seminr thus far.
Dear seminr,
I have thoroughly enjoyed using your hugely functional package for finishing up my thesis, however I have recently revised my model and have been unable to run the predict_pls function - this is quite important for my analyses.
The error I'm getting is, 'Error in ordered_data[trainIndexes, endogenous_items]: Can't subset columns with
endogenous_items
. x Subscriptendogenous_items
must be a simple vector, not a matrix'.I have tried cloning the pls_predict code and executing this, which worked before. However I haven't had much luck this time round. Any idea where or what the endogeneity is, or if I can change the matrix to a simple vector as it says?
I was wondering if you could highlight where I could amend this so I can finish my analyses? Just to note my seminr model is a mediated model - I also have an unmediated model, and have tried this too, but the error is different:
'error in sample.int(length(x), size, replace, prob): invalid 'size' argument.
Hope you can help with this, have had so much fun and success with seminr thus far.