5 Responses

1. Sarah

   August 8th, 2007 at 10:46 am

   1

   Try to make the bins smaller, and look for outliers. Data outside of 5 x (Standard Deviation).
   Then match those values with the corresponding dates and volatility. I think it should provide something good.
   For all I know you must have already done that.
   ~Sarah

2. Sarah

   August 8th, 2007 at 10:56 am

   2

   Are you sure the chances are .14% on any given day?

   .14/day
   => 14/100 days
   => 1/7 days
   That means this event happens ones every 7 days. I might be wrong - please correct me.
   ~Sarah

3. Tom

   August 8th, 2007 at 11:00 am

   3

   Ah yes you are right, I must’ve been drinking again. It should be 0.14% chance over 10,000 observations (trading days).

4. Sarah

   August 9th, 2007 at 8:47 am

   4

   Wow…
   Once in approximately 20 years…
   That’s significant.
   Do you know what happened last time this happened?
   ~Sarah

5. Tom

   August 9th, 2007 at 8:52 am

   5

   From Chuck Schwab:

   “For those with a short-term focus, here are some historical stats about market action following greater-than-2% down days for stocks, courtesy of Bespoke Investment Group: There have been 11 2% down days since early 2003. The average next-day change on the S&P 500 after a 2% down day has been 0.63%, with the index going up seven days and down four days. The average one-week change was 1.56% and the average one-month change was 2.53%. Most notably, on the prior two 2% down days thus far in 2007, the index went up the following day by an average of 0.61%.”