| Title: | Likelihood Criterion (LIC) Analysis for Laplace Regression Model |
| Version: | 3.0.0 |
| Date: | 2024-11-23 |
| Description: | Performs likelihood criterion analysis using the Laplace regression model to determine its optimal subset of variables. The methodology is based on Guo et al. (2023), LIC criterion for optimal subset selection in distributed interval estimation <doi:10.1080/02331888.2020.1823979>. |
| License: | MIT + file LICENSE |
| Encoding: | UTF-8 |
| RoxygenNote: | 7.3.2 |
| Imports: | stats, VGAM, dplyr, LaplacesDemon, relliptical, ggplot2, rlang |
| NeedsCompilation: | no |
| Packaged: | 2024-11-23 10:48:14 UTC; Lenovo |
| Author: | Guangbao Guo [aut, cre], Yaxuan Wang [aut] |
| Maintainer: | Guangbao Guo <ggb11111111@163.com> |
| Depends: | R (≥ 3.5.0) |
| Repository: | CRAN |
| Date/Publication: | 2024-11-28 12:00:02 UTC |
LLIC for Lre Model
Description
This function carries out an Laplace LIC analysis utilizing the Lre model.
Usage
LLIC(X, y, alpha, K)
Arguments
X |
Design matrix |
y |
Random response vector of observed values |
alpha |
Significance level |
K |
Number of subsets |
Value
A list containing the following components:
MUopt |
A vector of the means of the predictor variables in the optimal subset. |
Bopt |
A vector of the estimated regression coefficients from the final model fitted to the optimal subset. |
MAEMUopt |
The Mean Absolute Error (MAE) for the optimal subset. |
MSEMUopt |
The Mean Squared Error (MSE) for the optimal subset. |
opt |
Currently NULL, a placeholder for potential future use. |
Yopt |
A vector of the predicted values from the final model fitted to the optimal subset. |
Examples
set.seed(12)
library(VGAM)
X <- matrix(data = sample(1:3, 1200 * 5, replace = TRUE), nrow = 1200, ncol = 5)
b <- sample(1:3, 5, replace = TRUE)
e <- rlaplace(1200, 0, 1)
Y <- X %*% b + e
alpha <- 0.05
K <- 10
result <- LLIC(X, Y, alpha, K)
MUopt <- result$MUopt
Bopt <- result$Bopt
MAEMUopt <- result$MAEMUopt
MSEMUopt <- result$MSEMUopt
opt <- result$opt
Yopt <- result$Yopt
Data Processing for LLIC Analysis
Description
This function processes the data generated for the LLIC analysis, including filtering, mutation, and selection of specific columns.
Usage
data_pc(data)
Arguments
data |
A data frame containing the raw data generated for the LLIC analysis. |
Value
A data frame with the following columns:
X1 |
The filtered values of the original 'X1' column, keeping only rows where 'X1 <= 2'. |
X2 |
The original 'X2' column. |
X1_squared |
A new column containing the square of the 'X1' values. |
Examples
set.seed(12)
library(dplyr)
library(VGAM)
raw_data <- data.frame(
X1 = sample(1:3, 1200, replace = TRUE),
X2 = sample(1:3, 1200, replace = TRUE),
X3 = sample(1:3, 1200, replace = TRUE),
X4 = sample(1:3, 1200, replace = TRUE),
X5 = sample(1:3, 1200, replace = TRUE),
Y = rlaplace(1200, 0, 1)
)
processed_data <- data_pc(raw_data)
advanced_plotting_LLIC for LLIC
Description
This function visualizes the results of the LLIC analysis, including a comparison of actual and predicted values, and a bar chart of model coefficients.
Usage
plot_LLIC(X, Y, result)
Arguments
X |
Design matrix used in the LLIC analysis. |
Y |
Random response vector of observed values used in the LLIC analysis. |
result |
A list containing the results of the Laplace LIC analysis from the LLIC function. |
Value
A list containing two 'ggplot' objects:
Actual_vs_Pred |
A scatter plot comparing the actual vs predicted values. |
Coef |
A bar chart displaying the model's coefficients. |
Examples
set.seed(12)
library(VGAM)
library(rlang)
library(dplyr)
library(ggplot2)
X <- matrix(data = sample(1:3, 1200 * 5, replace = TRUE), nrow = 1200, ncol = 5)
b <- sample(1:3, 5, replace = TRUE)
e <- rlaplace(1200, 0, 1)
Y <- X %*% b + e
alpha <- 0.05
K <- 10
result <- LLIC(X, Y, alpha, K)
plot_LLIC(X, Y, result)
plots <- plot_LLIC(X, Y, result)
print(plots$Actual_vs_Pred)
print(plots$Coef)