StockFetcher Forums · Filter Exchange · PORTFOLIO SELECTION AND MANAGEMENT USING RISK/REWARD RATIOS<< 1 ... 35 36 37 38 39 ... 65 >>Post Follow-up
blumberg
27 posts
msg #105199
Ignore blumberg
3/1/2012 12:10:03 AM

Thanks, Kevin.

jackmack
333 posts
msg #105203
Ignore jackmack
3/1/2012 10:42:08 AM

Kevin
This is for the IRA filter - not the 401K version
Just curious what you make of your original filter from page 1.
If run today (as was yesterday - last day of the month) using EEM-IYW-SPY-SHY
EEM is the top pick now.
I read back through the thread and it states for IRA version one can use the
4 listed above and pick from the top one - again I am just wondering why one
would not use the EEM signal now for IRA purposes vs. IWM since this is what
the filter indicates.
Trying to remove emotion and my own geopolitical bias from the filter and use
just what it says - so in that case wouldn't the top pick now be EEM?

Kevin_in_GA
4,538 posts
msg #105207
Ignore Kevin_in_GA
3/1/2012 4:22:35 PM

I don't have an emerging markets option in my 401k, but have International as the only ex-US option (thus the use of EFA rather than EEM).

You can use any etf you want if you can buy it in your account. It should work just fine.

Kevin_in_GA
4,538 posts
msg #105222
Ignore Kevin_in_GA
3/2/2012 2:39:12 PM

At the end of each month, I usually re-run the timing optimizations for both the weekly and monthly ROCs using a rolling 5 year window.

Best weekly timing - still 14 weeks
Best monthly timing - still 3 months

Both say to be in IWM. IN fact, the top 5 best performing systems for both the weekly and monthly ROCs ALL are in IWM at the moment.

Just in case you're getting a little nervous (as I am) watching IWM continually underperform SPY over the last three weeks. Often that is a canary in a coal mine for a drop in the markets.

jgilberAZ
4 posts
msg #106387
Ignore jgilberAZ
modified
5/24/2012 9:23:27 PM

I'm having a hard time backtesting this.

Every time I do (and I can only choose 5 months range), I end up worse than the S&P.

=============================

Thought I should add more detail ...

I am using the filter Kevin posted as "this is the one I'm using" on page 30-31 (somewhere in that range).

Except, I have (only for the sake of backtesting) added this to the very end: rank below 2

That gives me one stock only.

For the backtest, I set it to only hold one position at a time, and only hold it for 4 days.
(Minimum hold=4, maximum hold=5)

That causes a sale and purchase once a week.

Sometimes, it sells and purchases the same stock at the same price. But, I don't see that as an issue.

I've tried it for several different time ranges, although I can only select 5 months at a time.

Here are the results from the latest backtest:

Basic Setup
Name: Weekly Shuffle
Approach Type: Long
Start Date: 01/23/2012
End Date: 05/23/2012
Benchmark Symbol: ^SPX

Exit Setup
Stop Loss: N/A
Profit Stop: N/A
Trailing Stop Loss: N/A
Minimum Holding Days: 4
Maximum holding days: 5


Trade Statistics
There were 17 total stocks entered. Of those, 17 or 100.00% were complete and or 0.00% were open.
Of the 17 completed trades, 10 trades or 58.82%resulted in a net gain.
Your average net change for completed trades was: -0.12%.
The average draw down of your approach was: -1.78%.
The average max profit of your approach was: 1.34%
The Reward/Risk ratio for this approach is: 0.87
Annualized Return on Investment (ROI): -6.17%, the ROI of ^SPX was: 0.76%.


Equity Summary
01/24/12 bought 1287.33 shares ($100000.00) of IWM at 77.68, remaining cash: $0.00
01/31/12 sold 1287.33 shares ($102548.92) of IWM at 79.66 for a gain of $2548.92
01/31/12 bought 1287.33 shares ($102548.92) of IWM at 79.66, remaining cash: $0.00
02/07/12 sold 1287.33 shares ($106346.55) of IWM at 82.61 for a gain of $3797.63
02/07/12 bought 1287.33 shares ($106346.55) of IWM at 82.61, remaining cash: $0.00
02/14/12 sold 1287.33 shares ($105445.42) of IWM at 81.91 for a loss of $901.13
02/14/12 bought 1287.33 shares ($105445.42) of IWM at 81.91, remaining cash: $0.00
02/22/12 sold 1287.33 shares ($105522.66) of IWM at 81.97 for a gain of $77.24
02/22/12 bought 1287.33 shares ($105522.66) of IWM at 81.97, remaining cash: $0.00
02/29/12 sold 1287.33 shares ($106269.31) of IWM at 82.55 for a gain of $746.65
02/29/12 bought 1287.33 shares ($106269.31) of IWM at 82.55, remaining cash: $0.00
03/07/12 sold 1287.33 shares ($101583.42) of IWM at 78.91 for a loss of $4685.89
03/07/12 bought 752.14 shares ($101583.42) of SPY at 135.06, remaining cash: $0.00
03/14/12 sold 752.14 shares ($105374.18) of SPY at 140.10 for a gain of $3790.76
03/14/12 bought 1270.18 shares ($105374.18) of IWM at 82.96, remaining cash: $0.00
03/21/12 sold 1270.18 shares ($105679.03) of IWM at 83.20 for a gain of $304.84
03/21/12 bought 752.06 shares ($105679.03) of SPY at 140.52, remaining cash: $0.00
03/28/12 sold 752.06 shares ($106115.22) of SPY at 141.10 for a gain of $436.19
03/28/12 bought 1266.29 shares ($106115.22) of IWM at 83.80, remaining cash: $0.00
04/04/12 sold 1266.29 shares ($104304.42) of IWM at 82.37 for a loss of $1810.80
04/04/12 bought 1266.29 shares ($104304.42) of IWM at 82.37, remaining cash: $0.00
04/12/12 sold 1266.29 shares ($100720.82) of IWM at 79.54 for a loss of $3583.60
04/12/12 bought 734.49 shares ($100720.82) of SPY at 137.13, remaining cash: $0.00
04/19/12 sold 734.49 shares ($101822.55) of SPY at 138.63 for a gain of $1101.74
04/19/12 bought 734.49 shares ($101822.55) of SPY at 138.63, remaining cash: $0.00
04/26/12 sold 734.49 shares ($102013.52) of SPY at 138.89 for a gain of $190.97
04/26/12 bought 734.49 shares ($102013.52) of SPY at 138.89, remaining cash: $0.00
05/03/12 sold 734.49 shares ($103078.53) of SPY at 140.34 for a gain of $1065.01
05/03/12 bought 734.49 shares ($103078.53) of SPY at 140.34, remaining cash: $0.00
05/10/12 sold 734.49 shares ($100390.30) of SPY at 136.68 for a loss of $2688.24
05/10/12 bought 734.49 shares ($100390.30) of SPY at 136.68, remaining cash: $0.00
05/17/12 sold 734.49 shares ($97584.54) of SPY at 132.86 for a loss of $2805.76
05/17/12 bought 1160.62 shares ($97584.54) of BND at 84.08, remaining cash: $0.00
05/23/12 sold 1160.62 shares ($97514.90) of BND at 84.02 for a loss of $69.64


Initial portfolio Size: $100000.00

At the end of the backtest:

Available Cash: $97514.90



Is there something wrong with my backtest, or is there something wrong with the filter and/or its application?



Thanks!

Kevin_in_GA
4,538 posts
msg #106417
Ignore Kevin_in_GA
5/25/2012 1:14:22 PM

Why would you only hold for 4 days? That is not at all how this system is executed. This is an investment strategy, not a trading strategy.



jgilberAZ
4 posts
msg #106419
Ignore jgilberAZ
modified
5/25/2012 1:27:34 PM

Because you said you re-assess and re-allocate weekly, and that was the only way I could figure out how to tell the backtest program to re-assess weekly.

If it re-assesses and selects the same stock as it held, that's the same as just keeping the stock, no? There are no commissions involved, and the sell/buy are the same.

I guess I don't know how to tell it "pick one stock, and re-assess the pick each week."

Is there a better way to backtest this?

=============================

I plugged the trades into excel, removed the sell/buy for the same stock ...

1/24/2012 bought 1287.33 shares ($100,000.00) of IWM at 77.68 remaining cash: $0.00
3/7/2012 sold 1287.33 shares ($101,583.42) of IWM at 78.91 gain of $1,583.42
3/7/2012 bought 752.14 shares ($101,583.42) of SPY at 135.06 remaining cash: $0.00
3/14/2012 sold 752.14 shares ($105,374.18) of SPY at 140.10 gain of $3,790.79
3/14/2012 bought 1270.18 shares ($105,374.18) of IWM at 82.96 remaining cash: $0.00
3/21/2012 sold 1270.18 shares ($105,679.03) of IWM at 83.20 gain of $304.84
3/21/2012 bought 752.06 shares ($105,679.03) of SPY at 140.52 remaining cash: $0.00
3/28/2012 sold 752.06 shares ($106,115.22) of SPY at 141.10 gain of $436.19
3/28/2012 bought 1266.29 shares ($106,115.22) of IWM at 83.80 remaining cash: $0.00
4/12/2012 sold 1266.29 shares ($100,720.82) of IWM at 79.54 loss of -$5,394.40
4/12/2012 bought 734.49 shares ($100,720.82) of SPY at 137.13 remaining cash: $0.00
5/17/2012 sold 734.49 shares ($97,584.54) of SPY at 132.86 loss of -$3,136.27
5/17/2012 bought 1160.62 shares ($97,584.54) of BND at 84.08 remaining cash: $0.00
5/23/2012 sold 1160.62 shares ($97,514.90) of BND at 84.02 loss of -$69.64

Comes out the same.


mahkoh
889 posts
msg #106422
Ignore mahkoh
modified
5/25/2012 4:50:57 PM

I think you came up with a clever "workaround" to backtest the filter with a weekly re-allocation setup and think your results would be accurate. However Kevin has backtested this to the bone and I would take his results to be genuine.
There is a performance graph on his website for the system

http://www.statisticalinvesting.com/investment-systems#!__investment-systems/vstc2=sharpe-401k-alllocation/vstc13=system-performance

One thing that struck me in this picture was the fact that it actually underperformed the index during the first years. It is when the market tanks and the system parks you in cash or bonds that it makes the big difference.

Edit: Come to think of it, this might actually be a good add-on to the filter: Reallocate on a regular basis,- be it monthly or weekly- but get your ass into cash the minute BND or AGG shows up on top. Curious as to whether this would improve performance..

mahkoh
889 posts
msg #106423
Ignore mahkoh
5/25/2012 6:13:03 PM

Trying to backtest this on a daily reallocation:


Entry Filter
symlist(iwm,vwo,vnq,moo,agg,tlt,uup,slv,gld)

add column weekly roc(13,1)
sort column 5 descending








Basic Setup
Name: 401k
Approach Type: Long
Start Date: 01/30/2012
End Date: 05/25/2012
Benchmark Symbol: ^SPX

Exit Setup
Stop Loss: N/A
Profit Stop: N/A
Trailing Stop Loss: N/A
Minimum Holding Days: N/A
Maximum holding days: 1

Extra Indicators
Entry Columns:
Show Performance After: after 2 days
after 5 days
after 10 days
after 25 days
after 40 days


Advanced Options
Selection Method: select by column 5
Entry Price: open
Conditional Entry: No
Exit Price: open
Maximum Trades Per Day: 1
Maximum Open Positions: 1
Maximum Selected Stocks: All
Close all OPEN positions: Yes

Soething strange about this, 82 trades, but all are GLD. Apparently it only picks the one with the highest price. What am I doing wrong??









jgilberAZ
4 posts
msg #106424
Ignore jgilberAZ
5/25/2012 11:59:09 PM

You have to make it evaluate all the stocks, but only return one.

To do that, rank them, then add a condition "rank below 2"

That way, it will only return the highest rated stock.


==================================================

symlist(iwm,vwo,vnq,moo,agg,tlt,uup,slv,gld)

set {roc13, weekly roc(13,1)}
add column roc13
sort column 5 descending

set {rIWM, IND(IWM, roc13)}
set {rVWO, IND(VWO, roc13)}
set {rVNQ, IND(VNQ, roc13)}
set {rMOO, IND(MOO, roc13)}
set {rAGG, IND(AGG, roc13)}
set {rTLT, IND(TLT, roc13)}
set {rUUP, IND(UUP, roc13)}
set {rSLV, IND(SLV, roc13)}
set {rGLD, IND(GLD, roc13)}

SET{RANK1A, COUNT(roc13 is above rIWM,1)}
SET{RANK1B, COUNT(roc13 is above rVWO,1)}
SET{RANK1C, COUNT(roc13 is above rVNQ,1)}
SET{RANK1D, COUNT(roc13 is above rMOO,1)}
SET{RANK1E, COUNT(roc13 is above rAGG,1)}
SET{RANK1F, COUNT(roc13 is above rTLT,1)}
SET{RANK1G, COUNT(roc13 is above rUUP,1)}
SET{RANK1H, COUNT(roc13 is above rSLV,1)}
SET{RANK1I, COUNT(roc13 is above rGLD,1)}


SET{ST1, RANK1A + RANK1B}
SET{ST2, RANK1C + RANK1D}
SET{ST3, RANK1E + RANK1F}
SET{ST4, RANK1G + RANK1H}

set {ST5, ST1 + ST2}
set {ST6, ST3 + ST4}
set {ST7, ST5 + ST6}
set {RANKTOT, ST7 + RANK1I}
SET{RANK, 9 - RANKTOT}

ADD COLUMN SEPARATOR
ADD COLUMN RANK {current rank}
ADD COLUMN SEPARATOR

rank below 2


StockFetcher Forums · Filter Exchange · PORTFOLIO SELECTION AND MANAGEMENT USING RISK/REWARD RATIOS<< 1 ... 35 36 37 38 39 ... 65 >>Post Follow-up

*** Disclaimer *** StockFetcher.com does not endorse or suggest any of the securities which are returned in any of the searches or filters. They are provided purely for informational and research purposes. StockFetcher.com does not recommend particular securities. StockFetcher.com, Vestyl Software, L.L.C. and involved content providers shall not be liable for any errors or delays in the content, or for any actions taken based on the content.


Copyright 2016 - Vestyl Software L.L.C.Terms of Service | License | Questions or comments? Contact Us
EOD Data sources: DDFPlus & CSI Data Quotes delayed during active market hours. Delay times are at least 15 mins for NASDAQ, 20 mins for NYSE and Amex. Delayed intraday data provided by DDFPlus