For the First Time, from Victor Niederhoffer
For the first time since October 2006 there were no declines of 1% or more in the month of March 2010. One hypothesizes that the number of such declines will increase from the three in April 2010. Indeed, one hypothesizes that a month with zero such declines like March 2010 is inordinately associated with the end of a bull market in Birinyian terms – assuming such exist.
Kim Zussman writes:
To clarify, these dates are for 21D periods which had one or more daily decline <-1% and were following a 21D period which had no declines <-1% (like March and April 2010). The first column after "Date" is count of declines <-1% in that period. Next col is that 21D period return. 3rd column is count of declines <-1% in the next 21D period, and the last column is the return for the next 21D period.
The last 4/5 mean returns for subsequent 21D periods have been negative:
Date count -1% 21d ret nxt 21-1% nxt21 ret
05/31/07 1 0.030 4 -0.018
01/30/07 1 0.001 1 -0.018
11/27/06 1 -0.005 0 0.032
04/28/06 1 0.006 4 -0.039
01/27/06 1 0.022 1 -0.002
10/26/05 3 -0.020 1 0.065
08/26/05 1 -0.031 0 0.009
11/26/04 1 0.051 1 0.026
07/29/04 2 -0.031 3 0.007
03/29/04 5 -0.020 3 0.000
01/28/04 1 0.030 0 0.015