Kevin_in_GA
4,599 posts
msg #100861
- Ignore Kevin_in_GA |
5/22/2011 9:34:47 PM
Again, someone wants to take a system designed and backtested on STOCKS and use it for options. This aproach may keep you in a specific investment for a year or more, or as little as a week. You never know. Given that reality, how would you factor in the time decay in options?
Obviously, I have heard this question many times before. Can it work with options? Sure. Will it work? Can't say. This system is designed for managing one's long-term investments (like in a 401k). Adapt it to other approaches at your own risk.
|
gmg733
788 posts
msg #100890
- Ignore gmg733 |
5/24/2011 4:12:15 PM
I'm an options trader and the first thing I thought was LEAPs. My second thought ,as I like to hedge, was a poor man's covered calls might be a better fit. Sell the near term, buy the long dated. Gives you upside potential and a nice hedge to the down side. Roll up and out as the instrument runs (for a credit that is).
You'll make money. But I also like the simplicity of Kevin's system.
|
seanban
22 posts
msg #101087
- Ignore seanban |
6/5/2011 1:30:54 PM
Based on the filter provided in this thread by Kevin and subsequent suggestions, I've done some back testing over the last 12 months.
Filter:
symlist(VB,VWO,TLT, DBC)
add column weekly roc(13,1) {13 week performance}
sort on column 5 descending
When a new signal is triggered, buy+sell at Open on first trading day of the week.
Overall result is 18.84%
I expect improvements can be made on this?
Suggestions are welcome.
EFT diversification Filter by Kevin_in_GA Backtest results from 11/6/2010 Profit/Loss % = 18.84
Signal Date Symbol Weeks Active Close Entry Date Exit Date Entry Price Exit Price Profit/Loss (%)
3/6/11 TLT 96,33 6/6/3011
27/5/11 VWO 1 48.3 28/5/11 6/6/2011 49.11
20/5/11 TLT 3 95.32 9/5/2011 30/5/2011 94.9 96.18 1.33
29/4/11 DBC 15 31.9 24/1/11 1/5/2011 27.89 29.03 3.93
13/1/11 VB 2 74.33 10/1/2011 24/1/2011 72.78 72.39 -0.54
31/12/10 DBC 1 27.55 3/1/2011 10/1/2011 27.83 27.34 -1.79
12/11/10 VB 7 67.54 15/11/2010 3/1/2010 67.88 73.38 7.50
3/9/10 VWO 10 42.65 6/9/2010 15/11/2010 42.39 47.45 10.66
27/8/10 TLT 1 105.35 30/8/2010 6/9/2010 106.21 105.09 -1.07
20/8/10 VWO 1 41.68 23/8/2010 30/8/2010 41.83 41.01 -2.00
13/8/10 TLT 1 102.29 16/8/2010 23/8/2010 104.21 105.65 1.36
6/8/10 VWO 1 42.64 9/8/2010 16/8/2010 42.96 41.32 -3.97
11/6/10 TLT 8 100.48 14/6/2010 9/8/2010 96.32 99.74 3.43
|
Kevin_in_GA
4,599 posts
msg #101088
- Ignore Kevin_in_GA
modified |
6/5/2011 2:36:34 PM
Here are all of the trades signaled using a modification of your ETF list and a 14 week ROC (which is what I use personally, and had mentioned this earlier in this thread).
TLT Long
05/16/2011 94.7
06/03/2011 96.56
1.84%
IWM Long
05/09/2011 83.19
05/16/2011 82.94
-0.42%
DBC Long
02/28/2011 29.69
05/09/2011 29.01
-2.41%
IWM Long
02/14/2011 82.13
02/28/2011 82.63
0.49%
DBC Long
01/31/2011 28.17
02/14/2011 28.74
1.90%
IWM Long
11/22/2010 71.86
01/31/2011 77.85
8.20%
EEM Long
09/07/2010 41.47
11/22/2010 45.89
10.51%
TLT Long
06/01/2010 93.26
09/07/2010 101.7
8.89%
Assuming you roll 100% of your equity into each trade as the system requires, I get the return for the past year at (1.089 * 1.105 * 1.082 * 1.019 * 1.005 * 0.976 * 0.996 * 1.002) = 29.9% return.
Running these specific ETFs might have changed the ROC period, but in fact it still seems that 14 weeks is the best. I suggested 13 weeks in the first filter because most 401k accounts have this info readily available for each of their funds.
Kevin
|
seanban
22 posts
msg #101090
- Ignore seanban |
6/5/2011 3:29:44 PM
Indeed. I did miss the ROC (14) reference. Thanks for your performance figures. No one can complain with these returns.
Sean
|
Kevin_in_GA
4,599 posts
msg #101697
- Ignore Kevin_in_GA |
7/18/2011 12:24:24 PM
Just following up on this, given the unusually high volatility over the past 4-6 weeks (massive ramp up, choppiness, and now correction). The relative strength model here has kept you in bonds since 5/16.
TLT Long
05/16/2011 94.70
07/18/2011 95.30
0.63%
Not as good as it was 6 weeks ago, but still profitable. Compared to IWM (down -1.95%), SPY (-2.76%), or EFA (-5.69%) over the same time frame it looks pretty good.
Kevin
|
duke56468
683 posts
msg #101712
- Ignore duke56468
modified |
7/18/2011 9:26:18 PM
Kevin....... I'm probably doing something wrong as usual,but I get at least 2 whipsaws during that time one july 1 IWM and one june 3 EEM
|
duke56468
683 posts
msg #101765
- Ignore duke56468 |
7/21/2011 8:19:29 PM
Kevin_in_GA
1,958 posts
msg #101697
- Ignore Kevin_in_GA 7/18/2011 12:24:24 PM
Just following up on this, given the unusually high volatility over the past 4-6 weeks (massive ramp up, choppiness, and now correction). The relative strength model here has kept you in bonds since 5/16.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Kevin....is this the filter you used? I get 2 whipsaws during that time with it. EEM for a loss and IWM for a gain.
|
jackmack
334 posts
msg #102186
- Ignore jackmack |
8/11/2011 1:42:50 PM
Kevin
On the 5-20 post you indicated you used other set of filters on another software set up.
Any chance I could get an idea of what that would be or is?
Thanks
|
jackmack
334 posts
msg #102187
- Ignore jackmack |
8/11/2011 2:42:33 PM
Kevin - I know this is a crazy thought but have you ever looked at each day of the week as the start date?
What I mean is instead of buying on Monday - have you ever looked at other days of the week to exit one ETF
and start a new position in another? Does this skew the results - better/worse?
Just curious.
Thank you
|