# [R-SIG-Finance] [R-sig-finance] Confusing result with AIC and ACF

RON70 ron_michael70 at yahoo.com
Tue Jan 20 14:44:29 CET 2009

```Hi, I have a strange problem with following dataset :

-0.075851693
-0.046125504
-0.009117161
0.025569817
0.034882743
0.073671497
0.063805297
0.062306796
0.072343820
0.058354121
-0.007635359
0.086790779
0.085487789
0.113577103
0.021293381
0.089423068
0.090485998
0.128847827
0.011859335
0.058794744
0.065909368
0.020887431
0.085387467
0.097375525
0.108981417
0.044289044
0.071428571
0.052430556
0.056307049
0.041957314

If I see the ACF then, it seems that this values are random. However next I
fit two regressions and get the AIC values :

1. regression of y (above series) on : y[-1], y[-2], y[-3], y[-4] and x[-1],
x[-2], x[-3], x[-4]

2. regression of y on y[-1], y[-2], y[-3] and x[-1], x[-2], x[-3], x[-4]

Here y[-p] is the p-th lagged value. and x is some other variable.

AIC for those regressions are : -6.673 and -6.636. This means 4-th lag of y
has some explanatory power on y. Therefore this result is coming confusing
when I compare with ACF figures [which tells y is random]

Can ppl here suggest me what inference I should make on y?

Regards,

--
View this message in context: http://www.nabble.com/Confusing-result-with-AIC-and-ACF-tp21563337p21563337.html
Sent from the Rmetrics mailing list archive at Nabble.com.

```