Talk:Partial autocorrelation function

???
This could certainly use some expansion. — Preceding unsigned comment added by 65.199.51.66 (talk) 16:26, 3 November 2008

Certainly. — Preceding unsigned comment added by 200.12.179.249 (talk) 16:29, 23 February 2009

Simplification
Please use a diagram to show the way people does. Jackzhp (talk) 03:23, 10 February 2010 (UTC)

something wrong?
in the description paragraph, "Given a time series z_t, the partial autocorrelation of lag k is the autocorrelation between z_t and z_{t+k} with the linear dependence of z_{t+1} through to z_{t+k-1} removed; equivalently, it is the autocorrelation between z_t and z_{t-k} that is not accounted for by lags 1 to k-1, inclusive.". I feel it is not right. Jackzhp (talk) 15:05, 3 June 2011 (UTC)

Is Durbin-Levinson Algorithm the same as Levinson Recursion?
In Shumway & Stoffer, the equation in the Calculation section is called the "Durbin-Levinson Algorithm" but I've seen other places use "Levinson-Durbin" or even "Levinson Recursion." I'm not sure if these are referring to the same thing especially since the Durbin-Levinson Algorithm can be found from a series of linear equations that can be turned into matrices being multiplied. I've previously linked the wiki page for Levinson recursion but I'm going to remove it in favor of just "Durbin-Levinson Algorithm" for now. This isn't necessarily my deepest subject of expertise so I might be making a dumb mistake. If anyone wants to chime in with their knowledge please do. Moon motif (talk) 03:28, 12 July 2022 (UTC)