# 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.

264 questions
Filter by
Sorted by
Tagged with
90 views

### Obtaining Residual Sum of Squares from a large OLS problem using a naive sequential approach - why doesn't it work?

Suppose we have an OLS problem with a large number of predictors: $Y = X_1 + X_2 + \cdots + X_p$. I want to obtain its RSS. I don't need to know the regression coefficients or individual residuals, ...
19 views

### Can I use clustering to compare diseases, after partial least squares correlation?

I have two datasets from patients with different diseases, which contain multiple measures of regional brain function and performance on different memory tasks. The memory tasks used in each dataset ...
18 views

### Why am I getting negative components with my custom NIPALS algorithm

I've recently been learning about the Nonlinear Iterative Partial Least Squares (NIPALS) algorithm for computing the principal components of a dataset. I am trying to code a NIPALS class from scratch ...
34 views

### How to interpret PLS-MGA’s results based on the significance of the corresponding group specific path coefficients

As a more general follow-up question to Elisa’s post on “[h]ow to interpret regression coefficients” when carrying out a PLS-structural-equation-modeling multigroup analysis (PLS-MGA), when comparing ...
17 views

### So significant serial mediating effect, but there is a significant mediating effect when the individual mediators are explored?

The statistical analysis used was PLS-SEM using WarpPLS.I have four variables, with 1 IV (W), 1 DV (Z), and 2 mediators (X and Y) I was able to find that when run through W -> X -> Y -> Z the ...
1 vote
13 views

### References for SIMPLS - Partial Least Squares

I am currently learning partial least squares analysis. I understand the idea, and the method I know to get PLS is NIPALS. Since I am currently writing a paper about this subject, I want to learn ...
1 vote
17 views

### PLSR: trait vs spectroscopic data gives very low R2 on plsr model in R

here is the sample data. I have spectroscopy data as X-variables (from X1 to X80) and corresponding Y variable. I need to run plsr model in R using "pls" package. There are two sheets. In ...
1 vote
9 views

### Could I remove LOCs of an HOC to have the compositional invariance?

So, I am testing a theory, proposed in the literature, and I want to do a multigroup analysis on it. One of the HOCs(reflective) of the model has three LOCs(all reflective); when I run the tests on ...
33 views

### How to predict Y using PLS-Canonical model given the X matrix?

Recently I've been learning Partial Least Squares (PLS), and learned that there are some variations of PLS, namely PLS-Canonical, PLS-SVD, PLS2 and PLS1 (mainly from the User Guide of PLS classes in ...
15 views

### PLS model on wide dataset with 60 samples and 120 variables

Data health for PLS modeling Hi, I am working on a manufacturing data that is fairly new (only has 60 batches produced so far) dataset size is 60 observations of 150 variables and I am building a PLS ...
1 vote
54 views

### Reflective-formative higher-order construct path issue in PLS-SEM

I have a reflective-formative higher-order construct that combines seven reflective first-order constructs. I validated this higher-order construct in a study using the PLS-SEM method in which all ...
16 views

### partial least squares structural equation modeling

I am creating a model and analyzing the relationships between the variables using the partial least squares structural equation. I have some questions regarding this method What do negative values of ...
14 views

### Sparse Partial Least Square: Variables in Components

I had 53 variables that I wanted to associate with response so I have performed sparse formulation of PLS and have one component which is optimally associated with response variable. This component is ...
25 views

### Preprocessing of spectroscopy data for PLSR: do I need to normalize the data for every wavelength?

I want to apply a partially least square regression on spectroscopy data to model a chemical content of my probe. So, every wavelength of the spectrum serves as one variable in the model. Doing some ...
16 views

### Canonical covariance for PLS in R

My question is quite simple, I don't have the answer though. I am working on some benchmark analysis to compare multivariate analyses, such as canonical correlation analysis or partial least squares. ...
42 views

### Choosing Which latent variables to display and analyze

Attached is a plot of the R-squared for each additional latent variable (LV) included in a PLSR. When choosing which LVs to display on a biplot (and ultimately determining which factors best explain ...
38 views

### high feature correlation but good OLS prediction

I have regression results where unconstrained OLS is near optimal - out of sample scores are almost the best when compared to some other constrained regression models. Although the ratio of number of ...
110 views

### Comparison of RMSE (root mean squared error) values

I want to see if my models work better univariate or multiple. But how can I do this? Normally I train the model, calculate the RMSE/MSE on the test data and compare these values. Now I trained the ...
13 views

### PLS and LDA analogies

I need some information to clarify some concept about PLS, LDA. LDA is able to decompose the independent variable to maximize the classes separation. the approach used is to develop a PCA on the ...
155 views

### Partial least squares component selection and overfitting

I am using partial least squares (PLS) to build a model using highly correlated data. I run a cross validation and get two well performing models: RMSE 1.39 and R2 0.76 which uses 8 components RMSE 1....
76 views

### Number of Principal Components in Principal Component Regression (PCR) with pcr and cross-validation implementation in R

I have a dataset with 401 predictors of 60 samples. It's from the Matlab Statistics and Machine Learning Toolbox: spectra.mat. I wanted to perform PCR on this dataset und afterwards determine the ...
200 views

### Cross validation of PLS-DA model

I have already asked this question in StackOverflow (https://stackoverflow.com/questions/74632176/how-can-i-cross-validate-the-classification-model), however, I would like to share it here as well, ...
481 views

### Reasoning behind variable importance in projection (VIP score)?

I have been using partial least squares regression to do latent variable modeling of a few data sets. Now I am wanting to look at variable importance but am getting stumped on the reasoning behind the ...
17 views

### PLS reporting correlation

I am very very new to PLS and haven't spent much time reading theory, but I have been nudged by my supervisor to use it to report some statistics for my thesis. I am currently using XLSTAT, and have 4 ...
47 views

### Is it sound to use PLS components in a model trained with the same response?

I have some high-dimensional spectral data I want to use in modeling plant productivity using a supervised model like random forest. I want to use the model for inference as well as for prediction in ...
117 views

### Why does X @ coef_ + intercept_ does not equal Y_pred for sklearn PLSRegression?

I performed partial least squares regression using Python's sklearn.cross_decomposition.PLSRegression using the example data in the sklearn docs. I am surprised that X @ coef_ + intercept_ does not ...
66 views

### In using PLS for feature selection, is directionality between X and Y necessarily implied?

In systems epidemiology such as in metabolomics, sometimes we are interested in identifying putative biomarkers of intake e.g., intake of a food item. So we analyse the metabolome to investigate ...
86 views

### Is it possible to predict on new data using PLS SEM?

Using the seminr package in R, I have fitted a model based on PLS SEM with several exogenous (latent) variables and one endogenous latent variable (ELV) measured by ...
1 vote
200 views

### How to work with Proportion of variance explained in PLS Regression?

I have some very basic questions regarding PLS. I ran PLS using SPSS on small dataset. n=312, Dependent variable=1; Independent Variables=9. Here is the output of Proportion of Variance Explained. As ...
69 views

### Regression in data with one group, having just zeros as outcome

I have a data set, consisting of positive and negative patients (virus infection). If the patient is negative, it has 0 as outcome (y), if it is positive it has a positive value, up to 100. The input (...
49 views

### R-square vs. NFI?

I ran a path model (no latent variables) in smartPLS3. It's not a complicated model. But after the analysis was computed, I checked the model fit measures. R-squares are small (all of them < .3), ...
56 views

### Transformation of data

I have a question regarding transformation of data. I have handled some data with both negative and positive elements by using the transformation: log(Y+1-min(Y)) which is all good. The problem is ...
1 vote
385 views

### How to convert SMART-PLS structural coefficients from standardized (correlation) to unstandardized?

I need use smartPLS but i also have prediction purpose (coefficients should also act like b-coefficients of regression). How to do this in smartPLS, since all coefficients are standardized?
1 vote
149 views

1 vote
123 views

### What is the PLSR implementation in sklearn?

Having a look to the source code from sklear implementation of PLSRegression I see two differences between what they cite as ...
116 views

### Variation on partial least squares regression (PLSR) that does not assume linear relationships between variables

Partial least squares regression (PLSR) is a statistical technique that allows you to predict multiple response variables from multiple predictor variables. It works by essentially running separate ...
79 views

### Multicollinearity and case:predictor ratio problems

I have 444 cases and 60ish predictors that suffer from collinearity. The predictors fall into three categories (vol, thickness and demographics). I would prefer to subdivide my cases into 4 (age) ...