Predict using the Polynomial Regression Model — predict_polynomial

This function predicts the response variable using the fitted Polynomial Regression model.

predict_polynomial_regression(model, new_data, x_vars, degree = 2)

Arguments

model: A list containing the fitted coefficients and the design matrix of the Polynomial Regression model
new_data: A data frame containing the new predictor variable values
x_vars: A character vector of column names corresponding to the predictor variables
degree: The degree of the polynomial (default = 2)

Value

A vector of predicted response variable values

Examples

# Generate sample data
set.seed(123)
n <- 100
p <- 10
data <- data.frame(matrix(runif(n * p, -1, 1), n, p))
colnames(data) <- paste0("x", 1:p)
data$y <- data$x1^2 + data$x2^2 + rnorm(n, sd = 0.1)

# Fit the Polynomial Regression using the function
fit <- fit_polynomial_regression(data, x_vars=paste0("x", 1:p), y_var="y", degree=2)

# Generate new data
new_data <- data.frame(matrix(runif(10 * p, -1, 1), 10, p))
colnames(new_data) <- paste0("x", 1:p)

# Predict the response variable using the fitted model
predictions <- predict_polynomial_regression(fit, new_data, x_vars=paste0("x", 1:p), degree=2)
print(predictions)
#>         [,1]
#> 1  0.6709962
#> 2  0.0695770
#> 3  0.4651418
#> 4  0.6782351
#> 5  0.1150972
#> 6  0.8299452
#> 7  0.8411522
#> 8  0.3750593
#> 9  1.0416450
#> 10 0.9291266