Correlation and Partial Autocorrelation
• lcorr(vx, vy)—Returns a vector giving the lagged sample correlation of vectors vx and vy at each possible lag.
• plcorr(v)—Returns a vector giving the partial autocorrelation of the vector v at each lag.
Arguments
• vx and vy are real-valued vectors of the same length.
• v is a real-valued vector.
Additional Information
• For both functions, the returned vector has the same length as the inputs.
• The lcorr function makes use of the FFT algorithm and zero-padding, which may introduce small real or imaginary components for values near 0.
• The kth component of lcorr measures the linear relationship between vx and the values of vy lagged by k units. Hence, lcorrk is referred to as the sample correlation at lag k.
• The plcorr function is used to estimate the order of the model that can best fit an autoregressive time series, and to help compute the model parameters.