swReg: Perform forward stagewise regression
In marjoleinF/swReg: Perform forward stagewise regression

Description Usage Arguments Details Value Examples

swReg performs forward stagewise regression.

1	swReg(X, Y, stepsize = 0.1, standardizeY = TRUE)

`X`	Matrix of predictor variables.
`Y`	Outcome variable (vector).
`stepsize`	numeric. Value with which the (un)standardized coefficients are updated in each iteration (a.k.a. epsilon).
`standardizeY`	logical. Should the response be standardized prior to application of the algorithm? If `TRUE`, stepsize can be interpreted as a correlation. If `sd(Y) > 1`, setting `standardizeY = FALSE` will increase the number of steps needed for convergence.

The function performs incremental forward stagewise regression, as described in Hastie, Tisbhirani & Friedman (2009). Hastie, T., Tibshirani, R., & Friedman, J. (2009). The Elements of Statistical Learning.

The function returns a list with the following elements: unstandardized.coef = bUnstand, standardized.coef = b, iteration = iteration, stepsize = stepsize (epsilon), coef.path = dataframe with standardized coefficient values for each predictor variable, at each stage or iteration of the algorithm data = list with original data (X andY)

## Example using Boston housing data:
 library(MASS)
 X <- as.matrix(Boston[,-14])
 Y <- Boston$medv
   
 ## Run forward stagewise regression:
 swReg1 <- swReg(X, Y, stepsize = .01)
 swReg2 <- swReg(X, Y, stepsize = .01, standardizeY = FALSE)