I am a user of your ARFIMA 1.01 module for ox. I just found a suspected bug in the ARFIMA 1.01. I specified the d parameter equal to 0 by using "arfima.FixD(0)". And get the following result. What make me confused is that the d parameter is not equal to 0 as I hoped. It seems "arfima.FixD(0)" fails to fix the d parameter. Can anybody knidly tell me what's wrong with this matter? Regards and thanks, Du Jun ---- Maximum likelihood estimation of ARFIMA(1,0,1) model ---- The estimation sample is: 1 - 2896 The dependent variable is: VOL Coefficient Std.Error t-value t-prob d parameter 0.491549 0.01148 42.8 0.000 AR-1 0.787656 0.1063 7.41 0.000 MA-1 -0.821159 0.09781 -8.40 0.000 Constant -13.7209 74.32 -0.185 0.854 Trend 0.0378687 0.009326 4.06 0.000