calcAvgAcc: Average Accuracy between Predicted and Actual Thermal...

In marcelschweiker/comfort_R: Models and Equations for Human Comfort Research

Average Accuracy between Predicted and Actual Thermal Sensation Vote

Description

calcAvgAcc calculates the average accuracy between predicted thermal sensation votes and actual obtained sensation votes

Usage

calcAvgAcc(ref, pred)

calcavgacc(ref, pred)

AvgAcc(ref, pred)

avgacc(ref, pred)

Arguments

ref	a numeric item or vector containing categorical actual thermal sensation votes coded from -3 'cold' to +3 'hot'
pred	a numeric item or vector containing categorical predicted thermal sensation votes coded from -3 'cold' to +3 'hot'

Value

calcAvgAcc returns a single value presenting the average accuracy between actual and predicted thermal sensation votes.

Note

The outcome heavily depends on the distribution of actual votes, i.e. in case most of the actual votes are in the same category, e.g. 'neutral', the average accuray is very high due to the fact that for the other categories the number of TRUE negative predicted votes is high as well.

Author(s)

Marcel Schweiker. Further contribution by Shoaib Sarwar.

References

Sokolova and Lapalme (2009) <doi:10.1016/j.ipm.2009.03.002>

See Also

calcTPRTSV, calcMeanBias

Examples

## Define data
ref <- rnorm(5) # actual thermal sensation votes
ref <- cutTSV(ref)

pred <- rnorm(5) # predicted thermal sensation votes
pred <- cutTSV(pred)

calcAvgAcc(ref, pred)

marcelschweiker/comfort_R documentation built on Feb. 22, 2024, 9:04 p.m.