plot - ways for cross correlation in R -
/ code> to plot it , But two plots are not the same. Agree with the BayPont plot of plot
ccf before acf , while the second plot by ccf does not agree with the Daamotom plot ACF . I wonder if I miss something? Thanks! CCF (x [, 1], x [, 2]) CCF (x [, 2], x [, 1])
par (mfrow = c (2,1) / Code>
ACF (x)
ACF measures the correlations of a time series against its own long. CCF measures co-relation between two time series at different times. That appears to be the first conspiracy by the CCF (and the left plot plot of the ACF) is actually different if you remove the values, so I'm sure you will see it. If there is a high relation between the two time-series, then it is very likely that the ACF and CCF will look the same.
The first conspiracy of the CCF actually matches the ACF plus Top ACF joint downstream. Top left and right ACF plots express only the correlations of time-series interval.
If you enter multivarium repeatedly in the ACF function, it will return the auto-correlation plot (top-left and lower right) with two parts of the cross-correlation function.
Comments
Post a Comment