I'm computing the past performance of stock market trading stategies. 
A common trading rule is momentum, which selects stocks on the basis
of their past returns (high past returns = buy; low past returns =
sell).  I rely on a short holding period of 1 month only, so I
re-balance the portfolio at the end of each month.  This generates
high turnover, but lends itself to statistical analysis.  Since there
are no overlapping holding periods, traditional standard errors are
fine.  But I want to extend the holding period (say for 3 months)
which means the portfolios overlapping.  Researchers compute
Newey-West standard errors to account for this, but I haven't yet
found an explanation of the methodology that I can understand.  One
more thing - I don't have access to stats/econometrics software, only
Excel.  So how do I calculate Newey-West standard errors in Excel?