wite noise in time series

[Chiraz KARAMTI]

Hi every body, I'm using the portemanteau test for white noise and I did the correlograme for the residuals here are the results:

LAG AC PAC Q Prob>Q [Autocorrelation] [Partial Autocor]
-------------------------------------------------------------------------------
1 0.0662 0.0662 5.0957 0.0240 | |
2 -0.0055 -0.0099 5.1308 0.0769 | |
3 -0.0346 -0.0338 6.5213 0.0888 | |
4 0.0630 0.0681 11.151 0.0249 | |
5 -0.0105 -0.0201 11.281 0.0461 | |
6 -0.0160 -0.0147 11.581 0.0720 | |
7 -0.0132 -0.0064 11.785 0.1079 | |
8 -0.0042 -0.0088 11.805 0.1601 | |
9 0.0475 0.0503 14.445 0.1074 | |
10 0.0251 0.0197 15.185 0.1255 | |

. tsline cac

. wntestq e

Portmanteau test for white noise
---------------------------------------
Portmanteau (Q) statistic = 36.6599
Prob > chi2(40) = 0.6214



Which looks bizarre to me is that the null hypothesis of presence of autocorrelation is accepted until lag(6) whith Ljung-Box statistic. However, the whole white noise test accept the null hypothesis of white noise for all the dataset. Is this normal??? I'm really confused!

[Chiraz KARAMTI]

sorry this is better :

. corrgram e in 2/1160, lags(10)

-1 0 1 -1 0 1
LAG AC PAC Q Prob>Q [Autocorrelation] [Partial Autocor]
-------------------------------------------------------------------------------
1 0.0662 0.0662 5.0957 0.0240 | |
2 -0.0055 -0.0099 5.1308 0.0769 | |
3 -0.0346 -0.0338 6.5213 0.0888 | |
4 0.0630 0.0681 11.151 0.0249 | |
5 -0.0105 -0.0201 11.281 0.0461 | |
6 -0.0160 -0.0147 11.581 0.0720 | |
7 -0.0132 -0.0064 11.785 0.1079 | |
8 -0.0042 -0.0088 11.805 0.1601 | |
9 0.0475 0.0503 14.445 0.1074 | |
10 0.0251 0.0197 15.185 0.1255 | |


. wntestq e

Portmanteau test for white noise
---------------------------------------
Portmanteau (Q) statistic = 36.6599
Prob > chi2(40) = 0.6214

when I reduce the lag it reject H0 (white noise) :

. wntestq e, lags(4)

Portmanteau test for white noise
---------------------------------------
Portmanteau (Q) statistic = 11.1514
Prob > chi2(4) = 0.0249


I'm I missing somthing???