davesaint86
726 posts
msg #97736
- Ignore davesaint86 |
12/2/2010 7:58:31 PM
ETFReplay is now charging $29.99 per month.
Dave
|
Kevin_in_GA
4,599 posts
msg #97752
- Ignore Kevin_in_GA |
12/3/2010 6:57:40 AM
Yes, but is anyone actually using their site to determine trades? I know we all like it, but I use the TSI system (IWM is up 6.5% since it was selected on 10/24, and VWO closed out up 7+% from the open on 10/13, and AGG was up 2.9% from the open on 5/10 until 10/10).
That means with just these three trades, you are up 17.2% since the open on May 10th. In comparison the SPY is only up 5.8%
|
davesaint86
726 posts
msg #97754
- Ignore davesaint86 |
12/3/2010 8:36:10 AM
Kevin,
I mainly used the backtesting portion of the system. I'm very happy using TSI. In fact I subcribe to www.netsignal.com as a backup. I'm thinking about cancelling the subscription today even though it's only $10 per month. NetSignal has you out out of emerging markets, domestic stocks and international right now. What I like about ETFReplay and TSI is that you rotate into the highest TSI or RS (if you use ETFReplay). My 401k is up 13.10% this year. I wasn't able to rotate a few times as quickly as I wanted to becasue of the time restrictions that the plan states must be followed. Thanks Kevin for developing the filter.
Davesaint
|
duke56468
683 posts
msg #97761
- Ignore duke56468 |
12/3/2010 9:30:48 AM
Kevin...You may have answered this else where, but in your back-testing did you assume entry and exit at the end of day rather than open the following day. I think ETFreplay used EOD. It can make a difference. Thanks for all your help.
|
Kevin_in_GA
4,599 posts
msg #97764
- Ignore Kevin_in_GA |
12/3/2010 10:18:35 AM
I have looked at both. To be honest, I have based my trading on the assumption that you exit from the current ETF at the close on the Friday the weekly TSI signal says to exit, and you enter the new trade at the opening on the following Monday.
In my real life execution of this, there is a delay of one or two days in the execution of any reallocation between funds (I use the ETFs as proxies for the different funds available for me). This is because I am using a Federal TSP plan for this trading account, which is notoriously slow. In my other accounts, the reallocations are executed at the close of business on the day the request is placed.
|
mahkoh
1,065 posts
msg #98656
- Ignore mahkoh |
1/19/2011 5:42:14 PM
401K filter says replace IWM with MOO
|
Kevin_in_GA
4,599 posts
msg #98661
- Ignore Kevin_in_GA |
1/19/2011 9:12:53 PM
Yes, but you should wait until the end of the week (Fridays). The weekly TSI is not finalized until then, and all of the backtesting was on finalized data, not mid-week.
I would look more closely (and this is what I have been doing lately) at the simple 13 week roc:
This is based on the work of Mebane Faber, and in my backtesting it did quite well. Currently still indicating that IWM is the right call. I look at both of these indicators and for quite a while now they have been signaling IWM (I'm up over 9% on this call since 11/13, even after today's 2.5% drop).
|
mahkoh
1,065 posts
msg #98928
- Ignore mahkoh |
2/5/2011 1:24:21 PM
Both 401K and the variation with ROC say: "sell IWM, buy MOO".
|
TraderMojo
16 posts
msg #98963
- Ignore TraderMojo |
2/8/2011 6:02:02 PM
Kevin,
Awesome filter! How does the backtesting performance of your weekly TSI compare to the ETFreplay methodology? I'm deciding whether to pay up for ETFreplay.
|
Kevin_in_GA
4,599 posts
msg #98979
- Ignore Kevin_in_GA |
2/9/2011 6:08:53 AM
Both approaches work, and have merit. The ETFReplay approach uses both performance and volatility rankings to make their selection, where the TSI or ROC approaches shown here use only performance.
This is why I have been requesting additional functions from SF such as a rank function or one that automatically calculates the Sharpe ratio for a user-specified time period. So far no progress on either front.
In my opinion, ETFReplay provides a valuable service worth the money. I personally use the methods listed here so I have not subscribed yet (and may never subscribe - not sure if I need to given that I am comfortable with this approach).
While ETFReplay and I use different means, the end is the same - to use relative strength as a guide for investment.
Please also check out www.ETFscreen.com - great site that uses relative strength as well, is currently free, and let's you set up personal portfolios. Backtesting is not nearly as good as ETFReplay, but their thinking, approach, and the volume of data they provide for free is worth evaluating.
Kevin
|