Feb

28

Measured yesterday's SPX closing price as a percentage decline of more than 4 standard deviations below the average one-day percentage change measured over the last 30 trading days. Examined the dates in history of like moves and the percentage-change T-days out. History shows a strong bounce averaging 4.3% by 2-days out, 9 of 10 winners and 1 no-change, t=3.2 at day-2.

Bernd Dittmann adds:

I continued Jay's counting, but instead of using confidence intervals, I looked at extreme values. Based on daily returns from Jan. 2, 1987 till today (4992 obs.), here are the left and right tails of the return distribution:

%return  <-% obs  normal  >+% obs

0.1           2184        2321       2418
0.5           1466        1865       1663
1                876        1337       1041
2                334         596          371
3                140         182          136
4                  69          50            61
5                  36           9             27
6                  23           1            15
7                  18           0             9
8                  12           0             5
9                   8            0             2
10                 6            0             2

What is clearly striking is that declines of 3% or more have been observed more frequently than 3%+ percent increases. If one were to use a normal distribution to describe Hang Seng daily returns (which is rejected at any level of significance), one would clearly underestimate the frequency of extreme returns. Which distribution would thus fit Hang Seng returns, and also its asymmetry in extreme values?

Larry Williams remarks: 

Traders should carefully note which stocks in the Dow were the least resistant to the selling pressures yesterday. An important subject, raised by Victor and Laurel a few weeks back.


Comments

Name

Email

Website

Speak your mind

2 Comments so far

  1. Bernd Dittmann on February 28, 2007 10:06 pm

    Only Taiwan’s (TSE) close slightly up! Might appear as rather ironic, but let’s not forget that their correlation is only 1.6% (based on a 05/01/2000 - present sample).

    By the way: TSE is not normally distributed. I’d thus be careful with sd’s. Easily rejected at less than 1 per cent.

  2. Tom Tarr on March 1, 2007 8:55 pm

    Market Performance 1 Day, 1 Week, 1 Month After a Large drop in the Dow.

    The market averages are the most bullish after 1 month, and in the case of the S&P 500, after a 400-point drop or greater, the S&P is positive 100% of the time, an average of 5%.

    After a 400 point drop in the Dow (6 Occurrences):

    1 Day later the Dow is up 83.33% of the time an average of 3.58%

    1 Day later the S&P 500 is up 83.33% of the time an average of 3.79%

    1 Day later the NASDAQ is up 66.67% of the time an average of 5.20%

    1 Week later the Dow is up 66.67% of the time an average of 4.23%

    1 Week later the S&P 500 is up 66.67% of the time an average of 3.91%

    1 Week later the NASDAQ is up 66.67% of the time an average of 5.46%

    1 Month later the Dow is up 83.33% of the time an average of 5.67%

    1 Month later the S&P 500 is up 100% of the time an average of 5.01%

    1 Month later the NASDAQ is up 83.33% of the time an average of 5.86%

    After a 200 point drop in the Dow (74 Occurrences) :

    1 Day later the Dow is up 56% of the time an average of 1.5%

    1 Day later the S&P 500 is up 64% of the time an average of 1.47%

    1 Day later the NASDAQ is up 58.67% of the time an average of 2.27%

    1 Week later the Dow is up 61.33% of the time an average of 3.08%

    1 Week later the S&P 500 is up 60.81% of the time an average of 3.28%

    1 Week later the NASDAQ is up 58.11% of the time an average of 5.05%

    1 Month later the Dow is up 69.33% of the time an average of 5.93%

    1 Month later the S&P 500 is up 74.67% of the time an average of 5.60%

    1 Month later the NASDAQ is up 66.67% of the time an average of 9.49%

    (Statistical Data Provided by LIM)

Archives

Resources & Links

Search