I'm trying to understand how ACF is actually calculated.

Let's say we have the following dataset. How would we calculate the ACF of YtYt-2.

t Yt
1 3
2 6
3 8
4 4
5 4
6 8

Question

I'm trying to understand how ACF is actually calculated.  

Let's say we have the following dataset.  How would we calculate the ACF of YtYt-2.  

t   Yt
1   3
2   6
3   8
4   4
5   4
6   8

amistre64 · Answer

hopefully you can better define what ACF stands for.

amistre64 · Answer

\bar y $$\bar y$$

anonymous · Answer

ACF is autocorrelation function

anonymous · Answer

The ACF function is below for reference:
$$\frac{\sum_{t=1+k}^{n} (Y _{t}-\bar Y)(Y _{t-k}-\bar Y) }{ \sum_{t = 1}^{n} (Y_{t} - \bar Y)^2}$$] \]

amistre64 · Answer

http://mathworld.wolfram.com/Autocorrelation.html autocorrelation is beyond the scope of my ken. can you make any sense of the wolfram documentation?

anonymous · Answer

Unfortunately not regarding the Wolfram documentation.  Thanks for taking a look.

amistre64 · Answer

good luck with it :)  there are plenty of smarter than mes about that may have a better head on their shoulders for this topic :)