This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: The higher the cov, the more confident we are about the future values …. just for linear autocorrelation  correlation defined for variables over different time points autocovariance (variance between x1 and x5 is autocovariance with 4 steps [over time]) auto  just measured over time correlation standardizes the covariance by making it on a scale from 1 to 1 ACF = autocorrelation function difference btwn xT and Xt1 is very small, second line is xt and x(t2) HOW TO CALCULATE ACF BY HAND Mean Absolute Error: MAE = mean(abs(Et)) [the average of the errors] Root mean square error: RMSE = root(mean(Et(squared))) Mean absolute percentage error: MAPE = mean(abs(pt)) where pt = 100Et / yt Least square estimate: sigma[yi  B0 B1(xi)](squared]...
View
Full Document
 Fall '13
 Lin
 Statistics, Standard Deviation, Variance, Probability theory, mean square error, generate random poisson

Click to edit the document details