Yes those are the links that have erroneous data. Not too big a deal for current calculation of NASI if you use the MEIKKA formula. However it might change the backtest numbers a little. Look for suspect &/or duplicate values on the following dates: 5/20/2002 7/4/2002 12/24/2002 12/31/2002 2/14/2003 4/17/2003 5/23/2003 8/29/2003 11/26/2003 12/24/2003 1/16/2004 4/9/2004 2/21/2005 5/30/2005 7/31/2001 5/2/1986