calc.mse: Calculate mean squared error (MSE) for vectors or matrices of...

In stas-g/funfun: Functions I use on a daily basis

Description

Calculate mean squared error (MSE) for vectors or matrices of observed and predicted values

Usage

calc.mse(obs, pred, rsq = FALSE)

Arguments

obs	A vector or a matrix/df (samples in rows, individual variables in columns) of observed values
pred	A vector or a matrix/df (samples in rows, individual variables in columns) of predicted values
rsq	Calculate R^2 (R-squared) instead of MSE? Defualt FALSE

Value

MSE or R^2 (single value if vectors are passed or a vector of length nrow(obs) if matrices/df are passed, one value for each obs-pred column pair)

Author(s)

N. F. Grinberg, [e-mail]

Examples

mod <- lm(mpg ~ ., data = mtcars)
summary(mod)$r.squared
calc.mse(mtcars$mpg, mod$fitted.values, rsq = TRUE)

stas-g/funfun documentation built on July 18, 2019, 12:58 p.m.