nega_gra: Function that calculate the negative gradient of given loss...
In StevenBoys/Ensemblelearn: Use Ensemble learning to better prediction.

Description Usage Arguments Value Examples

Function that calculate the negative gradient of given loss function

1	nega_gra(loss, x, y, coef, eps = 0.001)

`loss`	- the loss function used
`x`	- input independent variables x for the training
`y`	- the true dependent variable
`coef`	- the estimate of the coefficients in regression
`eps`	- the small increase used in the calculation of gradient

The negative gradient value of the given loss function.

x <- matrix(rnorm(4000), 200, 20)
beta <- rnorm(5)
y <- x[, 1:length(beta)] %*% beta + rnorm(200)
coef <- lm(y ~ -1 + x)$coefficients
loss <- mse
nega_gra(loss, x, y, coef)