rmse: Root Mean Square Error

In balexanderstats/bayesurvey: Bayesian Conjugate Prior Analysis of Survey Data

Description

This function calculates the root mean square error and checks that the idenifiers match. Note that if n=1, then the root mean square error is equivalent to the average error. This function can only be used for proportions.

Usage

rmse(predicted, actual)

Arguments

predicted	the predictions including the first column with a character vector with idenifiers to check that the rows of predicted and actual match
actual	the real result including the first column with a character vector

Value

the Root Mean Square Error

Examples

predict1 = data.frame(state = c("TX", "GA", "OR"), dem = c(.45, .49, .63), rep = c(.55, .51, .37))
actual1 = data.frame(state = c("TX", "GA", "OR"), dem = c(.42, .48, .62), rep = c(.58, .52, .38))
rmse(predict1, actual1)
predict2 = data.frame(state = c("OK", "MS", "WY"), dem = c(.42, .44, .3), rep = c(.56, .55, .7), other = c(0.02, 0.01, 0))
actual2 = data.frame(state = c("OK", "MS", "WY"), dem = c(.41, .38, .3), rep = c(.55, .58, .69), other = c(0.04, 0.04, 0.01))
rmse(predict2, actual2)

balexanderstats/bayesurvey documentation built on Sept. 20, 2020, 11:40 a.m.