Computes Various Model Fit Measures

logLik(m) computes the log likelihood of a cm object, SSE(m) computes the sum of squared errors, MSE(m) computes the mean squared error.

Usage

# S3 method for cm
logLik(object, newdata = NULL, ...)

MSE(x)

RMSE.cm(x)

SSE(x, ...)

SSE(x)

Arguments

...

other arguments (ignored)

x

a cm object

Value

A number measuring the goodness of fit between predictions and observed data.

Details

If a model predicts several values the error measures use the first column of predictions to compute the errors. For example, if the predictions are pr(x) and pr(z), the sum of squared errors is based on the data - pr(x).

See also

Other fit measures for cognitive models: AICc.cm(), MSE.cm()

Other fit measures for cognitive models: AICc.cm(), MSE.cm()

Other fit measures for cognitive models: AICc.cm(), MSE.cm()

Other fit measures for cognitive models: AICc.cm(), MSE.cm()

Examples

MSE(M)     # 0.1805
#> Error in MSE(M): object 'M' not found

D <- data.frame(x = 1, y = 1:1, z = 0:1)
M <- bayes_beta(y ~ x + z, D, fix = "start")
#> Error in bayes_beta(y ~ x + z, D, fix = "start"): could not find function "bayes_beta"
# If you want, look at the predictions
# predict(M)

SSE(M)     # 0.361
#> Error in SSE(M): object 'M' not found