Akaike information criterion (AIC) [AIC]

Share This

“A goodness of fit measure of forecast error based on the squared difference between observed and predicted Y values. In time series analysis, it measures the error from the autoregressive component and becomes smaller as the AR variance decreases. It is used to compare various models. In general, AIC= 2k + n*ln(residual sum of squares/n), where k is the number of parameters and is the likelihood value. Also see Schwarz criterian.”

« Back to Glossary Index Download Tooltip Pro
By |February 1st, 2019|Comments Off on Akaike information criterion (AIC) [AIC]