Ridge regression | R Documentation |
Regularisation via ridge regression is performed.
ridge.reg(target, dataset, lambda, B = 1, newdata = NULL)
target |
A numeric vector containing the values of the target variable. If the values are proportions or percentages, i.e. strictly within 0 and 1 they are mapped into R using log( target/(1 - target) ). |
dataset |
A numeric matrix containing the variables. Rows are samples and columns are features. |
lambda |
The value of the regularisation parameter λ. |
B |
Number of bootstraps. If B = 1 no bootstrap is performed and no standard error for the regression coefficients is returned. |
newdata |
If you have new data and want to predict the value of the target put them here, otherwise, leave it NULL. |
There is also the lm.ridge command in MASS library if you are interested in ridge regression.
A list including:
beta |
The regression coefficients if no bootstrap is performed. If bootstrap is performed their standard error appears as well. |
seb |
The standard erorr of the regression coefficients. If bootstrap is performed their bootstrap estimated standard error appears. |
est |
The fitted values if no new data are available. If you have used new data these will be the predicted target values. |
Michail Tsagris
R implementation and documentation: Michail Tsagris mtsagris@uoc.gr
Hoerl A.E. and R.W. Kennard (1970). Ridge regression: Biased estimation for nonorthogonal problems. Technometrics, 12(1): 55-67.
Brown P. J. (1994). Measurement, Regression and Calibration. Oxford Science Publications.
ridgereg.cv
#simulate a dataset with continuous data dataset <- matrix(runif(100 * 30, 1, 100), nrow = 100 ) #the target feature is the last column of the dataset as a vector target <- dataset[, 10] dataset <- dataset[, -10] a1 <- ridge.reg(target, dataset, lambda = 0.5, B = 1, newdata = NULL) a2 <- ridge.reg(target, dataset, lambda = 0.5, B = 100, newdata = NULL)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.