Tuesday, October 26, 2010

Vix and S&P 500 Rise Two Consecutive Days

Today the S&P 500 and VIX both rose for the 2nd consecutive day [S&P 500 technically only rose 0.02 points, but I'm still counting it.] Usually these two move in opposite directions. The VIX normally spikes when the S&P 500 declines, and vice versa. I wanted to test what typically happens when they both rise for 2 straight days. Is there any bias, either upside or downside?


When VIX and S&P 500 both rise 2 consecutive days, buy S&P 500 on close; sell 'n' days later. $100,000 per trade; 1990 to present.
Net Profit
# Trades
# of winners
% of Winners
Max. Trade % DD
Avg % P/L
W. Avg. Profit
L. Avg. Loss
Profit Factor
Payoff Ratio
1
  (12,566.76)
62
25
40.32
-3.05
-0.20
     448.14
     (642.44)
0.47
0.70
2
  (19,265.29)
60
30
50
-5.97
-0.32
     633.83
  (1,276.00)
0.50
0.50
3
  (19,692.31)
59
28
47.46
-8.45
-0.33
  1,005.33
  (1,543.27)
0.59
0.65
4
  (23,547.88)
57
28
49.12
-8.45
-0.41
  1,206.57
  (1,976.96)
0.59
0.61
5
  (20,697.18)
57
30
52.63
-8.45
-0.36
  1,306.55
  (2,218.28)
0.65
0.59
6
    (7,493.80)
56
30
53.57
-8.45
-0.13
  1,527.80
  (2,051.06)
0.86
0.74
7
    (5,609.55)
55
25
45.45
-8.54
-0.10
  1,766.66
  (1,659.20)
0.89
1.06
8
    (8,421.66)
53
27
50.94
-8.54
-0.16
  1,583.81
  (1,968.64)
0.84
0.80
9
    (8,082.12)
53
26
49.06
-8.54
-0.15
  1,872.15
  (2,102.15)
0.86
0.89
10
    (4,378.13)
52
32
61.54
-8.74
-0.08
  1,615.32
  (2,803.42)
0.92
0.58
11
      7,817.74
51
31
60.78
-8.74
0.15
  1,802.40
  (2,402.83)
1.16
0.75
12
    26,008.53
51
32
62.75
-8.74
0.50
  1,958.07
  (1,928.93)
1.71
1.02
13
    30,221.44
50
34
68
-8.74
0.59
  2,106.04
  (2,586.50)
1.73
0.81
14
    25,116.09
50
33
66
-8.74
0.49
  2,165.84
  (2,726.85)
1.54
0.79
15
    10,311.92
49
30
61.22
-11.19
0.21
  2,388.93
  (3,229.26)
1.17
0.74
16
    10,375.49
48
29
60.42
-11.52
0.21
  2,610.19
  (3,437.89)
1.16
0.76
17
      5,080.52
47
30
63.83
-12.50
0.11
  2,361.30
  (3,868.14)
1.08
0.61
18
      2,572.94
47
30
63.83
-12.50
0.05
  2,335.53
  (3,970.17)
1.04
0.59
19
      3,384.45
46
30
65.22
-12.50
0.07
  2,305.57
  (4,111.41)
1.05
0.56
20
    (1,178.65)
45
30
66.67
-12.50
-0.03
  2,204.10
  (4,486.78)
0.98
0.49


It appears that there is a definite downside bias over the next 10 days, especially during the 1st week after the event. After 10 days, the edge appears to reverse and display an upside bias for the next 6 to 7 days. There is a sweet spot between 12 and 14 days that I cannot explain; it may be an aberration.

Bottom line, this study is telling us to be cautious over the next 2 weeks, with the probability for lower prices very strong over the next 5 days. Intermediate term, there appears to be an upside bias, but it is not very consistent and probably not worth acting upon.

Footnotes:
Test was run in Amibroker. The portfolio buys exactly $100K each time. The data is from Yahoo Finance.The results are frictionless. There is no slippage or commissions; I just wanted to test a theory.

2 comments:

Bill Luby said...

Nice work, Chris. Based on what I have seen so far, I am going to find some room on my blogroll to squeeze you in.

Kudos and welcome to the blogosphere!

-Bill

Chris said...

Bill,

I am way late to responding here. Thanks for welcoming to the new world of blogging. You are an excellent source of information and inspiration. Thanks!

-Chris

Post a Comment