| Title: | Self-Validated Ensemble Models with Elastic Net Regression |
|---|---|
| Description: | Implements Self-Validated Ensemble Models (SVEM, Lemkus et al. (2021) <doi:10.1016/j.chemolab.2021.104439>) using Elastic Net regression via 'glmnet' (Friedman et al. <doi:10.18637/jss.v033.i01>). SVEM averages predictions from multiple models fitted to fractionally weighted bootstraps of the data, tuned with anti-correlated validation weights. Also implements the randomized permutation whole model test for SVEM (Karl (2024) <doi:10.1016/j.chemolab.2024.105122>). Code for the whole model test was taken from the supplementary material of Karl (2024). Development of this package was assisted by 'GPT o1-preview' for code structure and documentation. |
| Authors: | Andrew T. Karl [cre, aut]
|
| Maintainer: | Andrew T. Karl <[email protected]> |
| License: | GPL-2 | GPL-3 |
| Version: | 1.0.3 |
| Built: | 2024-11-21 09:10:28 UTC |
| Source: | CRAN |
The SVEMnet package implements Self-Validated Ensemble Models (SVEM) using Elastic Net (including lasso and ridge) regression via glmnet. SVEM averages predictions from multiple models fitted to fractionally weighted bootstraps of the data, tuned with anti-correlated validation weights.
SVEMnetFit an SVEMnet model using Elastic Net regression.
svem_significance_testPerform a whole-model significance test for SVEM models.
predict.svem_modelPredict method for SVEM models.
plot.svem_modelPlot method for SVEM models.
coef.svem_modelPlot method for SVEM models.
Development of this package was assisted by GPT o1-preview, which helped in constructing the structure of some of the code and the roxygen documentation. The code for the significance test is taken from the supplementary material of Karl (2024) (it was handwritten by that author).
Maintainer: Andrew T. Karl [email protected] (ORCID)
Gotwalt, C., & Ramsey, P. (2018). Model Validation Strategies for Designed Experiments Using Bootstrapping Techniques With Applications to Biopharmaceuticals. JMP Discovery Conference. https://community.jmp.com/t5/Discovery-Summit-2018/Model-Validation-Strategies-for-Designed-Experiments-Using/ta-p/73730
Karl, A. T. (2024). A randomized permutation whole-model test heuristic for Self-Validated Ensemble Models (SVEM). Chemometrics and Intelligent Laboratory Systems, 249, 105122. doi:10.1016/j.chemolab.2024.105122
Lemkus, T., Gotwalt, C., Ramsey, P., & Weese, M. L. (2021). Self-Validated Ensemble Models for Design of Experiments. Chemometrics and Intelligent Laboratory Systems, 219, 104439. doi:10.1016/j.chemolab.2021.104439
Xu, L., Gotwalt, C., Hong, Y., King, C. B., & Meeker, W. Q. (2020). General Applications of the Fractional-Random-Weight Bootstrap. Technometrics, 62(3), 345–358. doi:10.1080/00031305.2020.1731599
Ramsey, P., Gaudard, M., & Levin, W. (2021). Accelerating Innovation with Space Filling Mixture Designs, Neural Networks and SVEM. JMP Discovery Conference. https://community.jmp.com/t5/Abstracts/Accelerating-Innovation-with-Space-Filling-Mixture-Designs/ev-p/756841
Ramsey, P., & Gotwalt, C. (2018). Model Validation Strategies for Designed Experiments Using Bootstrapping Techniques With Applications to Biopharmaceuticals. JMP Discovery Conference - Europe. https://community.jmp.com/t5/Discovery-Summit-Europe-2018/Model-Validation-Strategies-for-Designed-Experiments-Using/ta-p/51286
Ramsey, P., Levin, W., Lemkus, T., & Gotwalt, C. (2021). SVEM: A Paradigm Shift in Design and Analysis of Experiments. JMP Discovery Conference - Europe. https://community.jmp.com/t5/Abstracts/SVEM-A-Paradigm-Shift-in-Design-and-Analysis-of-Experiments-2021/ev-p/756634
Ramsey, P., & McNeill, P. (2023). CMC, SVEM, Neural Networks, DOE, and Complexity: It’s All About Prediction. JMP Discovery Conference.
This function calculates the percentage of bootstrap iterations in which each coefficient is nonzero.
## S3 method for class 'svem_model' coef(object, ...)## S3 method for class 'svem_model' coef(object, ...)
object |
An object of class |
... |
other arguments to pass. |
Invisibly returns a data frame containing the percentage of bootstraps where each coefficient is nonzero.
Development of this package was assisted by GPT o1-preview, which helped in constructing the structure of some of the code and the roxygen documentation. The code for the significance test is taken from the supplementary material of Karl (2024) (it was handwritten by that author).
Plots actual versus predicted values for an svem_model using ggplot2.
## S3 method for class 'svem_model' plot(x, plot_debiased = TRUE, ...)## S3 method for class 'svem_model' plot(x, plot_debiased = TRUE, ...)
x |
An object of class |
plot_debiased |
Logical; if |
... |
Additional arguments passed to |
This function creates an actual vs. predicted plot for the SVEM model. If plot_debiased is TRUE and debiased predictions are available, it includes them in the plot.
**Plot Features:**
**Actual vs. Predicted Points:** Plots the actual response values against the predicted values from the SVEM model.
**Debiased Predictions:** If available and plot_debiased is TRUE, debiased predictions are included.
**Ideal Fit Line:** A dashed line representing perfect prediction (slope = 1, intercept = 0) is included for reference.
A ggplot object showing actual versus predicted values.
Development of this package was assisted by GPT o1-preview, which helped in constructing the structure of some of the code and the roxygen documentation. The code for the significance test is taken from the supplementary material of Karl (2024) (it was handwritten by that author).
Plots the Mahalanobis distances for the original and permuted data from multiple SVEM significance test results.
## S3 method for class 'svem_significance_test' plot(..., labels = NULL)## S3 method for class 'svem_significance_test' plot(..., labels = NULL)
... |
One or more objects of class |
labels |
Optional character vector of labels for the responses. If not provided, the function uses the response variable names. |
This function creates a combined plot of the Mahalanobis distances (d_Y and d_pi_Y) for the original and permuted data from multiple SVEM significance test results. It groups the data by response and source type, displaying original and permutation distances side by side for each response.
**Usage Notes:**
Use this function to compare the significance test results across multiple responses.
The plot shows original and permutation distances next to each other for each response.
A ggplot object showing the distributions of Mahalanobis distances for all responses.
Development of this package was assisted by GPT o1-preview, which helped in constructing the structure of some of the code and the roxygen documentation. The code for the significance test is taken from the supplementary material of Karl (2024) (it was handwritten by that author).
Generates predictions from a fitted svem_model.
## S3 method for class 'svem_model' predict(object, newdata, debias = TRUE, se.fit = FALSE, ...)## S3 method for class 'svem_model' predict(object, newdata, debias = TRUE, se.fit = FALSE, ...)
object |
An object of class |
newdata |
A data frame of new predictor values. |
debias |
Logical; default is |
se.fit |
Logical; if |
... |
Additional arguments. |
A debiased fit is output (along with the standard fit). This is provided to allow the user to match the output of JMP. https://www.jmp.com/support/help/en/18.1/?utm_source=help&utm_medium=redirect#page/jmp/overview-of-selfvalidated-ensemble-models.shtml. The debiasing coefficients are always calculated by SVEMnet(), and the predict() function determines whether the raw or debiased predictions are returned via the debias argument.
Predictions or a list containing predictions and standard errors.
Development of this package was assisted by GPT o1-preview, which helped in constructing the structure of some of the code and the roxygen documentation. The code for the significance test is taken from the supplementary material of Karl (2024) (it was handwritten by that author).
Performs a whole-model significance test using the SVEM framework, handling both continuous and categorical predictors.
svem_significance_test( formula, data, nPoint = 2000, nSVEM = 5, nPerm = 125, percent = 90, nBoot = 200, glmnet_alpha = c(1), weight_scheme = c("SVEM", "FWR"), debias = FALSE, objective = c("wSSE", "wAIC"), verbose = FALSE, ... )svem_significance_test( formula, data, nPoint = 2000, nSVEM = 5, nPerm = 125, percent = 90, nBoot = 200, glmnet_alpha = c(1), weight_scheme = c("SVEM", "FWR"), debias = FALSE, objective = c("wSSE", "wAIC"), verbose = FALSE, ... )
formula |
A formula specifying the model to be tested. |
data |
A data frame containing the variables in the model. |
nPoint |
The number of random points to generate in the factor space (default: 2000). |
nSVEM |
The number of SVEM models to fit to the original data (default: 5). |
nPerm |
The number of SVEM models to fit to permuted data for reference distribution (default: 125). |
percent |
The percentage of variance to capture in the SVD (default: 90). |
nBoot |
The number of bootstrap iterations within SVEM (default: 200). |
glmnet_alpha |
The alpha parameter(s) for glmnet (default: |
weight_scheme |
The weight scheme to use in SVEM (default: "SVEM"). Valid options are "SVEM" and "FWR". |
debias |
Logical; debiasing option passed to |
objective |
Character; the objective function to use in |
verbose |
Logical; if |
... |
Additional arguments passed to the underlying |
The 'svem_significance_test' function implements a whole-model test designed to gauge the significance of a fitted SVEM model compared to the null hypothesis of a constant response surface. This method helps identify responses that have relatively stronger or weaker relationships with study factors.
The test constructs standardized predictions by centering the SVEM predictions by the response mean and scaling by the ensemble standard deviation. A reference distribution is created by fitting the SVEM model to multiple randomized permutations of the response vector. The Mahalanobis distances of the original and permuted models are calculated using a reduced-rank singular value decomposition.
The R code to perform this test (using matrices of nSVEM and nPerm predictions) is taken from the supplementary material of Karl (2024).
A list containing the test results.
Development of this package was assisted by GPT o1-preview, which helped in constructing the structure of some of the code and the roxygen documentation. The code for the significance test is taken from the supplementary material of Karl (2024) (it was handwritten by that author).
Karl, A. T. (2024). A randomized permutation whole-model test heuristic for Self-Validated Ensemble Models (SVEM). Chemometrics and Intelligent Laboratory Systems, 249, 105122. doi:10.1016/j.chemolab.2024.105122
# Simulate data set.seed(0) n <- 21 X1 <- runif(n) X2 <- runif(n) X3 <- runif(n) y <- 1 + X1 + X2 + X1 * X2 + X1^2 + rnorm(n) data <- data.frame(y, X1, X2, X3) # Perform the SVEM significance test test_result <- svem_significance_test( y ~ (X1 + X2 + X3)^2 + I(X1^2) + I(X2^2) + I(X3^2), data = data, nPoint = 2000, nSVEM = 5, nPerm = 125, nBoot = 200 ) # View the p-value test_result$p_value test_result2 <- svem_significance_test( y ~ (X1 + X2 )^2 + I(X1^2) + I(X2^2), data = data, nPoint = 2000, nSVEM = 5, nPerm = 125, nBoot = 200 ) # View the p-value test_result2$p_value # Plot the Mahalanobis distances plot(test_result,test_result2)# Simulate data set.seed(0) n <- 21 X1 <- runif(n) X2 <- runif(n) X3 <- runif(n) y <- 1 + X1 + X2 + X1 * X2 + X1^2 + rnorm(n) data <- data.frame(y, X1, X2, X3) # Perform the SVEM significance test test_result <- svem_significance_test( y ~ (X1 + X2 + X3)^2 + I(X1^2) + I(X2^2) + I(X3^2), data = data, nPoint = 2000, nSVEM = 5, nPerm = 125, nBoot = 200 ) # View the p-value test_result$p_value test_result2 <- svem_significance_test( y ~ (X1 + X2 )^2 + I(X1^2) + I(X2^2), data = data, nPoint = 2000, nSVEM = 5, nPerm = 125, nBoot = 200 ) # View the p-value test_result2$p_value # Plot the Mahalanobis distances plot(test_result,test_result2)
Wrapper for 'glmnet' (Friedman et al. 2010) to fit an ensemble of Elastic Net models using the Self-Validated Ensemble Model method (SVEM, Lemkus et al. 2021). Allows searching over multiple alpha values in the Elastic Net penalty.
SVEMnet( formula, data, nBoot = 200, glmnet_alpha = c(0, 0.25, 0.5, 0.75, 1), weight_scheme = c("SVEM", "FWR", "Identity"), objective = c("wSSE", "wAIC"), ... )SVEMnet( formula, data, nBoot = 200, glmnet_alpha = c(0, 0.25, 0.5, 0.75, 1), weight_scheme = c("SVEM", "FWR", "Identity"), objective = c("wSSE", "wAIC"), ... )
formula |
A formula specifying the model to be fitted. |
data |
A data frame containing the variables in the model. |
nBoot |
Number of bootstrap iterations (default is 200). |
glmnet_alpha |
Elastic Net mixing parameter(s) (default is |
weight_scheme |
Weighting scheme for SVEM (default is "SVEM"). Valid options are "SVEM", "FWR", and "Identity".
"FWR" calculates the Fractional Weight Regression (Xu et al., 2020) and is included for demonstration;
"SVEM" generally provides better performance."Identity" simply sets the training and validation weights to 1. Use with |
objective |
Objective function for selecting lambda (default is "wSSE"). Valid options are "wAIC" and "wSSE". The "w" refers to "weighted" validation. |
... |
Additional arguments passed to the underlying |
The Self-Validated Ensemble Model (SVEM, Lemkus et al., 2021) framework provides a bootstrap approach to improve predictions from various base learning models, including Elastic Net regression as implemented in 'glmnet'. SVEM is particularly suited for situations where a complex response surface is modeled with relatively few experimental runs.
In each of the 'nBoot' iterations, SVEMnet applies random exponentially distributed weights to the observations. Anti-correlated weights are used for validation.
SVEMnet allows for the Elastic Net mixing parameter ('glmnet_alpha') to be a vector, enabling the function to search over multiple 'alpha' values within each bootstrap iteration. Within each iteration, the model is fit for each specified 'alpha', and the best 'alpha' is selected based on the specified 'objective'.
objective options:
"wAIC"Weighted Akaike Information Criterion. Balances model fit with complexity by penalizing the number of parameters. It is calculated as AIC = n \* log(wSSE / n) + 2 \* k, where wSSE is the weighted sum of squared errors, n is the number of observations, and k is the number of parameters. Typically used with weight_scheme="FWR" or weight_scheme="Identity"
"wSSE"Weighted Sum of Squared Errors. Selects the lambda that minimizes the weighted validation error without penalizing model complexity. While this may lead to models that overfit when the number of parameters is large relative to the number of observations, SVEM mitigates overfitting (high prediction variance) by averaging over multiple bootstrap models. Typically used with weight_scheme="SVEM"
weight_scheme options:
"SVEM"Uses anti-correlated fractional weights for training and validation sets, improving model generalization by effectively simulating multiple training-validation splits (Lemkus et al. (2021)). Typically used with objective="wSSE".
"FWR"Fractional Weight Regression as described by Xu et al. (2020). Weights are the same for both training and validation sets. This method does not provide the self-validation benefits of SVEM but is included for comparison. Typically used with objective="wAIC".
"Identity"Uses weights of 1 for both training and validation. This uses the full dataset for both training and validation, effectively disabling the self-validation mechanism. Use with objective="wAIC" and nBoot=1 to fit the Elastic Net on the AIC of the training data.
A debiased fit is output (along with the standard fit). This is provided to allow the user to match the output of JMP, which returns a debiased fit whenever nBoot>=10.? https://www.jmp.com/support/help/en/18.1/?utm_source=help&utm_medium=redirect#page/jmp/overview-of-selfvalidated-ensemble-models.shtml. The debiasing coefficients are always calculated by SVEMnet(), and the predict() function determines whether the raw or debiased predictions are returned via its debias argument.
The output includes:
**Model Output:**
The returned object is a list of class svem_model, containing the following components:
parms: Averaged coefficients across all bootstrap iterations.
debias_fit: The debiasing linear model fit (if applicable). This is a linear model of the form y ~ y_pred, used to adjust the predictions and reduce bias.
coef_matrix: Matrix of coefficients from each bootstrap iteration. Each row corresponds to a bootstrap iteration, and each column corresponds to a model coefficient.
nBoot: Number of bootstrap iterations performed.
glmnet_alpha: The Elastic Net mixing parameter(s) used. This is the alpha parameter from glmnet.
best_alphas: The best alpha values selected during the fitting process for each bootstrap iteration.
best_lambdas: The best lambda values selected during the fitting process for each bootstrap iteration.
weight_scheme: The weighting scheme used in SVEM. Indicates whether "SVEM", "FWR", or "Identity" weights were used.
actual_y: The response vector used in the model.
training_X: The predictor matrix used in the model.
y_pred: The predicted response values from the ensemble model before debiasing.
y_pred_debiased: The debiased predicted response values (if debiasing is applied). Adjusted predictions using the debias_fit model.
nobs: The number of observations in the dataset.
nparm: The number of parameters (including the intercept), calculated as ncol(X) + 1.
formula: The formula used in the model fitting.
terms: The terms object extracted from the model frame.
An object of class svem_model.
Development of this package was assisted by GPT o1-preview, which helped in constructing the structure of some of the code and the roxygen documentation. The code for the significance test is taken from the supplementary material of Karl (2024) (it was handwritten by that author).
Gotwalt, C., & Ramsey, P. (2018). Model Validation Strategies for Designed Experiments Using Bootstrapping Techniques With Applications to Biopharmaceuticals. JMP Discovery Conference. https://community.jmp.com/t5/Discovery-Summit-2018/Model-Validation-Strategies-for-Designed-Experiments-Using/ta-p/73730
Karl, A. T. (2024). A randomized permutation whole-model test heuristic for Self-Validated Ensemble Models (SVEM). Chemometrics and Intelligent Laboratory Systems, 249, 105122. doi:10.1016/j.chemolab.2024.105122
Lemkus, T., Gotwalt, C., Ramsey, P., & Weese, M. L. (2021). Self-Validated Ensemble Models for Design of Experiments. Chemometrics and Intelligent Laboratory Systems, 219, 104439. doi:10.1016/j.chemolab.2021.104439
Xu, L., Gotwalt, C., Hong, Y., King, C. B., & Meeker, W. Q. (2020). General Applications of the Fractional-Random-Weight Bootstrap. Technometrics, 62(3), 345–358. doi:10.1080/00031305.2020.1731599
Ramsey, P., Gaudard, M., & Levin, W. (2021). Accelerating Innovation with Space Filling Mixture Designs, Neural Networks and SVEM. JMP Discovery Conference. https://community.jmp.com/t5/Abstracts/Accelerating-Innovation-with-Space-Filling-Mixture-Designs/ev-p/756841
Ramsey, P., & Gotwalt, C. (2018). Model Validation Strategies for Designed Experiments Using Bootstrapping Techniques With Applications to Biopharmaceuticals. JMP Discovery Conference - Europe. https://community.jmp.com/t5/Discovery-Summit-Europe-2018/Model-Validation-Strategies-for-Designed-Experiments-Using/ta-p/51286
Ramsey, P., Levin, W., Lemkus, T., & Gotwalt, C. (2021). SVEM: A Paradigm Shift in Design and Analysis of Experiments. JMP Discovery Conference - Europe. https://community.jmp.com/t5/Abstracts/SVEM-A-Paradigm-Shift-in-Design-and-Analysis-of-Experiments-2021/ev-p/756634
Ramsey, P., & McNeill, P. (2023). CMC, SVEM, Neural Networks, DOE, and Complexity: It’s All About Prediction. JMP Discovery Conference.
Friedman, J. H., Hastie, T., & Tibshirani, R. (2010). Regularization Paths for Generalized Linear Models via Coordinate Descent. Journal of Statistical Software, 33(1), 1–22.
# Simulate data set.seed(0) n <- 21 X1 <- runif(n) X2 <- runif(n) X3 <- runif(n) y <- 1 + 2*X1 + 3*X2 + X1*X2 + X1^2 + rnorm(n) data <- data.frame(y, X1, X2, X3) # Fit the SVEMnet model with a formula model <- SVEMnet( y ~ (X1 + X2 + X3)^2 + I(X1^2) + I(X2^2) + I(X3^2), glmnet_alpha = c(1), data = data ) coef(model) plot(model) predict(model,data)# Simulate data set.seed(0) n <- 21 X1 <- runif(n) X2 <- runif(n) X3 <- runif(n) y <- 1 + 2*X1 + 3*X2 + X1*X2 + X1^2 + rnorm(n) data <- data.frame(y, X1, X2, X3) # Fit the SVEMnet model with a formula model <- SVEMnet( y ~ (X1 + X2 + X3)^2 + I(X1^2) + I(X2^2) + I(X3^2), glmnet_alpha = c(1), data = data ) coef(model) plot(model) predict(model,data)