Questions tagged [partial-least-squares]
A class of linear methods for modeling the relationship between two groups of variables, X and Y. Includes PLS regression.
204
questions
0
votes
2answers
32 views
PLSr: Generating predicted value using regression coefficient
I perform PLS with pls package in R using plsr function.
Why am I unable to get the same predicted Y value as when I use the ...
0
votes
1answer
22 views
Minimum number of obs. for machine learning and training/test sets?
Are there a minimum number of observations for ML techniques (classification, regression) in psychology/cognitive neuroscience? In particular for training and test datasets?
I found this article for ...
1
vote
0answers
11 views
Are there constraints on the ratio of dependent variables to independent variables for PLS regression?
I will begin with a disclaimer that while I understand the general underlying principles behind PLS, my linear algebra background is rather limited.
I have trouble with the details of constraints on ...
3
votes
1answer
84 views
RMSE behaviour during cross-validation of a PLS model - What does it mean?
In PLS regression, one of the approaches in selecting the number of components (Latent Variables, LV) is to perform cross-validation over a range of LV, and select the one with lower Root Mean Squared ...
0
votes
0answers
10 views
Applying PLSDA coefficients to a new dataset to classify [closed]
I have a trained model to classify leaves through its spectra by using a PLSDA (with caret package). My question is, is it possible to apply the coefficients of my model over a new spectra dataset?
0
votes
0answers
8 views
Partial least square
In the context of confirmatory factor analysis, structural equation modeling, & predictor space dimension reduction.
PLS is a supervised dimension reduction procedure, since it summarize the ...
0
votes
0answers
20 views
Comparison/Visualisation of Regression Methods
This question follows this question, in particular @amoeba's clarifying answer and the plot from the SAS documentation included. I'm especially interested in knowing
if $\mathbf{X}, \mathbf{Y}$ are ...
0
votes
1answer
41 views
PLS training by cross validation, selection of components and prediction
I know there are plenty of questions on this topic on the site, but I still can't work out how to choose the optimal number of components for the particularities of my PLS model.
I have 1800 ...
1
vote
1answer
17 views
Standardizing dataset before comparing slopes in regression
I have 3 different yield parameters data which was obtained from four different locations. Here is the distribution of dataset at all locations.
I want to determine which yield component is strongly ...
1
vote
0answers
17 views
Why is there an inconsistency between PLS projections and energy statistic output?
On R, I use the mixOmics package to run a repeated measures ("multilevel") Partial Least Squares analysis. I'm able to plot the 2-dimensional PLS projections such ...
0
votes
0answers
15 views
Treating seasonality in Partial Least Squares forecast
I've been looking for answers on this question but couldn't find concrete solutions so wanted to ask y'all.
I have been playing around trying to forecast an economic/financial-related indicator with ...
0
votes
0answers
7 views
heterotrait-monotrait analysis with PLSPM
I am using the PLSPM package in R and conducted discriminant analysis by applying the Fornell-Larcker criterion. I would also like to conduct heterotrait-monotrait (HTMT) analysis. Are there any tools ...
1
vote
0answers
51 views
How should predicted probabilities be interpreted in a binary classification model?
I ran a PLS-DA model with 10-fold cross-validation to classify data in 2 groups (using the Caret package in R). The predicted probabilities are close to 0.5 (the highest propbability is just 0.7). The ...
0
votes
0answers
31 views
Problem with Principal component (PCA) and Partial least squares (PLS) using R
I'm trying to reduce highly dimensional data with factor methods. I'm using Principal component analysis and Partial least squares.
From these methods I'm using the first component as a Common factor ...
1
vote
0answers
48 views
meaning of projection subspace in a PLSDA plot
I have a dataset with a handful of predictors and one output variable which is categorical and can only be C or N.
I am working ...
0
votes
0answers
22 views
How to validate PLS2 with R?
I have a PLS1 modell wich predicts single chemicals. Now I would like to run a PLS2 modell to predict two or more chemicals from a mixture. I programmed it in R and the modell runs, but i don´t find ...
0
votes
0answers
28 views
Interpretation of Nominal Scaling in PLSPM
I used nominal scaling in PLSPM in R and I am uncertain how to interpret the results. I made a minimal code example from the satisfaction data set and made up a nominal value for color (red, blue, ...
1
vote
0answers
25 views
How is this standard error obtained?
I am working through the exercises in Kuhn and Johnson's "Applied Predictive Modelling" and cannot reproduce one of their results in the exercises.
Looking at 4.3 we have
... find the number of ...
0
votes
1answer
143 views
How to interpret weights of a PLS SEM model
I made a PLS SEM model using smartPLS, consisting only of formative constructs. I managed to get weights out of the software, which all had excellent t values. The only thing is, I am not entirely ...
0
votes
1answer
58 views
When does partial least squares provide >1 component solutions?
I'm a beginner to using partial least squares analyses, so apologies if this question is a bit basic.
I've been trying out PLS models on my datasets and it usually says that a single component can ...
0
votes
0answers
9 views
What is an appropriate technique for large number of correlated predictor variables with interactions?
I want to allow the effects of a large number of continuous predictor variables to be different depending on which treatment group the individual is a part of.
If I had only three continuous ...
1
vote
1answer
45 views
Is this an example of where I shouldn't scale before doing PCA / PLS?
I'm working with NMR spectra (it's a common chemical test).
There are various peaks of the signal across a range of ppm values. I'm trying to relate the NMR spectra of various samples to a ...
0
votes
0answers
82 views
Partial Least squares regression - Variable Importance on Projection (VIP) method of selecting variables
I understand that partial least squares regression produces VIP scores for each predictor variable enabling variable selection (using a VIP threshold of >1).
Does this method account for collinearity ...
0
votes
0answers
39 views
Replicating partial least squares (NIPALS) results using ordinary least squares regression in Tensorflow?
I have multivariate variables that I want to regress to a single target label. For some reason, using partial least squares regression (projected to a single component) gives much better prediction ...
0
votes
0answers
25 views
Partial Least Square - choosing number of factors
I'm trying to implement PLS in R (using the package "PLS") to a time series consisting on realized variance of the S&P 500 and macroeconomic variables, however, I've notice they use cross-...
2
votes
1answer
285 views
Partial Least Squares regression - coefficients vs loadings
In partial least squares regression, what is the difference between the regression coefficients and the loadings for each independent variable in each component? Specifically, I understand in evety ...
2
votes
1answer
96 views
What is the origin of the PLS1 algorithm given on the PLS Wikipedia page?
The Wikipedia page for Partial Least Squares (PLS) gives an algorithm for the method which is uncited and for which I cannot find the source material. It appears to be very much simpler than most if ...
0
votes
0answers
84 views
What's the difference between loadings from partial least squares (PLS) regression and beta coefficients from multiple linear regression?
I have a set of independent variables (X1, X2, ..., X10) and I have run a PLS to find a combination of the X1, X2, ..., X10 that best predicts an outcome Y (a single-variable outcome).
As a result, I ...
0
votes
1answer
46 views
why does preprocessed test data change with change of calibration data in PLS-DA?
Why does preprocessed test data change when calibration dataset (and model based on that data) changes? I have spectral, normalized datasets, the preprocessing was 1. derivative + autoscale.
For ...
0
votes
0answers
57 views
Calculating Variable Importance for Feature Selection - PLSR
I have used the plsr() function in R (from the pls package) to predict a Y variable using many X variables (spectral bands) - and am wanting to calculate variable importance (ViP) to begin to reduce ...
1
vote
0answers
61 views
Is there a theoretical basis for using partial least squares with categorical responses
I am using what is called PLS-DA in JMP to find a model for predicting a categorical (Positive/Negative) response. The documentation says that the responses are simply coded as 0/1, thereby ...
1
vote
0answers
32 views
Partial Least Square vs Principle Component Regression
Is it the case where PLS, when compared to PCR with all things equal, generally gives lower bias but higher variance when regressed against a response Y, since PLS relates to/makes use of Y but PCR ...
3
votes
0answers
106 views
Partial Least Squares Regression : deflation of the Y matrix
I am digging deep into the PLSR algorithms and while I have found multiple flavours of if (different normalisations, SIMPLS,..), there is always something in the Y deflation that is throwing me off ...
0
votes
0answers
32 views
Is it appropriate/possible for me to use PLS regression for my problem?
Im currently writing a dissertation on the effect of cultural dimensions on technology acceptance. In order to collect data I have used a questionnaire with questions relating to technology acceptance,...
1
vote
0answers
50 views
PLS (Partial Least Squares) deflation and graphics
I have been working with pls for a little while now. I have a question in terms of the deflation of both the $X$ and $Y$ matrices.
In the literature I have found different methods over which deflation ...
3
votes
1answer
60 views
Formulating Partial Least Squares as minimizing squared error
The book chapter linked below (see section 4.3.1) lists a few formulations of partial least squares (PLS). The first two make sense to me and seem standard:
$$\underset{\mathbf{u}, \mathbf{v}}{\text{...
1
vote
0answers
21 views
How to compute/plot the contribution of each original descriptor in a final PLA regression model?
New to scikit-learn. I am using v 20.2. I am developing PLS regression models.I would like to know how important each of the original predictors/descriptors are in predicting the response. The ...
0
votes
0answers
74 views
Choosing number of factors in PLSR
Im confused about how many factors I should choose for my prediction model. I am using Unscrambler X to do PLSR. Unscrambler is supposed to suggest the optimal number of factors. It suggests 4 factors ...
1
vote
0answers
20 views
Will PLSR work on nonlinear dataset?
Im new to this topic and a bit confused. When I read about PLSR on the web I only see examples where the original plot shows a somewhat linear relationship in the data set, but in my case I have a ...
0
votes
0answers
33 views
Partial least squares and its variants
What is the difference between PLS, PLS-DA, PLS-PM and PLS-SEM? I could not find a single article that describes the differences.
1
vote
0answers
30 views
Screening data prior to PCA v. PLS
I have a very large time series matrix $X$, where the number of observations (rows) $n$ is much smaller than the number of input variables (columns) $p$. My aim is to use the information in $X$ to ...
0
votes
0answers
25 views
Non-linear multivariate analysis / partial least squares
When dealing with multivariate analysis with variables that have a mix of linear and non-linear relationships (example: two variables are linearly related but one of these maybe non-linear with ...
0
votes
0answers
27 views
3% explained variance in PLSR
My dataset contains 3 inputs and 1 output with 36000 samples. With PCA analysis I can explain 100% of the variance with 2 components. But when I do PLSR I only get 3% explained variance. Ive only done ...
2
votes
2answers
564 views
Interpret reuslts of PLS regression coefficients
I have performed PLS regression using sklearn library (python 2.7) over three types of soil (PLS model per soil type) and I plotted the regression coefficients, but ...
0
votes
0answers
54 views
I'd like to do regression using canonical correlation analysis
I got two multidimensional datasets, X and Y. I thought I build the model, which explains the relationship between two datasets, using canonical correlation analysis (CCA). The first correlation ...
1
vote
0answers
55 views
What is the point of using PRESS instead of RMSECV?
What is the point of using predicted residual sum of squares (PRESS) instead of root-mean-squared-error-of-cross-validation(RMSECV)? In many books, especially in the area of chemometrics, the authors ...
0
votes
1answer
50 views
Finding correlation between 2 predictors and a response
I've created a dataset, where the response, y, is related to the predictors X1, X2, by the formula: y = 2X1 + 5X2.
If we look at correlating y with X1, and then y with X2, we get the following:
And ...
1
vote
0answers
26 views
Partial least square different maximization programs
PLS regression is a regression method based based on latent variables in order to handle collinearity or violation of full rank assumption in linear regression.
Latent variables called components are ...
0
votes
1answer
26 views
Confirmatory Factor Analysis - Variance Parameters - Error Variance Estimates
I am referring to a confirmatory factor analysis output. I am curious to know how do we interpret the "Variance Parameters" Output containing the "error variance estimates" if the predictors are ...
1
vote
0answers
31 views
Inference using model with low goodness of fit.
Assuming a model is correctly specified, would it be appropriate to draw inferences based on a model with a low Goodness of Fit (~0.15)?
Of course, using such a model to make predictions is likely ...
