Reputation: 2233
I would like to check if there is a correlation between "birds" & "wolfs" in different lags.Getting the correlation value is easy but how can I address the lag issue ( I need to check the correleation value for 1:4 lags )? The output that I look for is a data table that contains the lag value and the related correlation value.
df <- read.table(text = " day birds wolfs
0 2 21
1 8 4
2 2 5
3 2 4
4 3 6
5 1 12
6 7 10
7 1 9
8 2 12 header = TRUE)
Output(not real results): Lag CorValue
0 0.9
1 0.8
2 0.7
3 0.9
Upvotes: 2
Views: 4246
Reputation: 3678
If you do this :
corLag<-ccf(df$birds,df$wolfs,lag.max=max(df$day))
it will return this :
Autocorrelations of series ‘X’, by lag
-8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8
-0.028 0.123 -0.045 -0.019 0.145 -0.176 -0.082 -0.126 -0.296 0.757 -0.134 -0.180 0.070 -0.272 0.549 -0.170 -0.117
the first row is the lag, the second is the correlation value. you can check that cor(df$birds,df$wolfs)
is indeed equal to -0.296
Upvotes: 4