|
|
|
--------------------
|
ETF Sector Strategy
The goal of this article is to demonstrate a relatively simple method of investing in ETFs. A trend following system was chosen to keep funds invested in the currently dominant sectors of the market. A monthly review cycle was chosen to minimize the time commitment of the strategy. This isn't hyper-day-trading or anything like that, but it is active management. In fact, as you will soon see, the system only switches funds about 4 times per year. Because we don't want all of our eggs in one basket we will always have our funds divided among two sectors of the market. More precisely, two sectors or styles. This project began with the family of 9 Select Sector SPDR ETFs. A fund family created to divide the S&P 500 stocks into independent market sectors. Being sub-sets of the S&P 500 stocks and equity weighted, these ETF's all tend to be dominated by large cap stocks. But large cap is not always where you want to be so the style-based Russell iShares funds were added to expand the options. In hindsight we could probably have just added the small-cap Russell ETFs, but we included all 12. That gives us 21 ETFs to screen each month, covering the major market sectors and styles. The next question to be answered was how to sort them. We could do some complex screen, but we only have a short history to backtest so the decision is made to keep it simple and use the trailing 6 month return. On this site that equates to the past 126 trading days. Each month we'll calculate the total return for these 21 funds over the prior 126 days (ending on the last day of the prior month) and sort the ETFs by this return in descending order. The first month we'll buy the top two positions, but in subsequent months we'll hold a position as long as it is in the top 6. This keeps us in the top quarter to third of our universe and reduces fund switches. If we do exit a position, or both of them, we'll buy the top funds we don't already own. If one posiiton is closed out we'll just roll that money into the new position. If both are closed we'll divide proceeds between the new funds. This is an imperfect rebalancing method but saves on transaction fees. ETFs are relatively new to the investment scene with most being created after 2000. The Select Sector family in our test began trading in late 1998 while the Russell ETFs began in early 2000. Since we need 6 months of trading history we'll begin our short backtest on January 2, 2001. You can see in Table 1 how the funds ranked for the first 12 months of our backtest.
The first month we buy the top two positions, XLF and XLU, and we hold these until they drop below position 6 in our screen. Following the information in table 1 you will see that as of the first of March XLF has dropped to position 8 triggering a sell of that holding and a buy of IWN, which was the top position we did not already hold. Then by May XLU has dropped to the number 7 position and is replaced in our portfolio by XLB. IWN and XLB remain in the top 6 positions throughout the remainder of the period shown in Table 1 and well into 2002. Table 2 shows all the trades in our backtest along with the dates and returns for those trades. The test began on 1/2/2001 and ended on 8/31/2005. You can see that 10 of the 16 trades were positive with 9 outperforming the SPY during the same time period. At least as important to me is that during periods of significant market movement our model seemed to outperform. When the market was up we were up more, and when the market was down we were either up or down less.
The CAGR of this system was 15.7% over this time period while the market returned an annualized 0.4%. Not to say that such performance will continue but it is nice to start knowing your system has the potential to significantly outperform the market. More stats are in Table 3.
Month by month performance data for the model and the SPY are in Tables 4 and 5. As you can see from this data our simple model has outperformed the broad market in 37 of 56 months and for every year of our test period.
As the graph below shows this model has nearly doubled in value during a time when the market was basically flat. That performance did not come evenly, however, as even this model was not immune to the bear market of 2002. During that year and again in 2003 this model experienced declines of over 20% from it's prior valuation peaks. But even then it's valuation stayed well above what the SPY would have provided. 
This trend following system appears to perform well on sector funds and the system is simple and straight forward. The 6-month returns can be calculated in a spreadsheet or can be obtained from this website. If you have any questions or comments feel free to post them using the form below, or email me using the link at the bottom of the page. I plan to post a system using other etf groups in the future. If you have a preference of family or other suggestion let me know. Disclaimers: Note that these results should be considered hypothetical, that the closing prices used may not have been achievable in real trading, and that no transaction fees are included. Also recognize that past performance does not necessarily predict future performance and therefore you should perform your own due diligence before following this or any other investment strategy. © 2005 H.M Todd
|
|
Current Comments
143 comments so far (post your own)There have been a few requests for updated stats and charts. I had hoped to post a good bit of data with some refinements to the strategies but since everything is not ready Ill just post updated strategy results. See sectorstrategyupdate.php.
Posted by hmTodd on Wednesday, 03.3.10 @ 12:13pm | #933
February Update - This sector strategy gained 5.2% in February while the SPY gained 3.1%. For the year this strategy has gained 2.8% while the broad market has lost 0.6%. The holdings are unchanged at MidCap Value(IWS) and Consumer Discretionary(XLY).
Hugh
Posted by hmTodd on Monday, 03.1.10 @ 12:47pm | #930
Hugh, will you be updating the performance graphs for the sector & international strategies vs the SPY. I believe the drawdowns will be helpful especially for 2008. Thanks
Posted by Earl on Sunday, 02.14.10 @ 07:47am | #929
A lot of information here. It is a little hard to interpret. What years in the past 10 has the strategy underperformed the S&P?. What was the performance in 2009. thank you.
Posted by daniel hogan on Friday, 02.5.10 @ 19:36pm | #927
January Update - This sector strategy lost 2.2% in January while the SPY lost 3.6%. The holdings are unchanged at MidCap Value(IWS) and Consumer Discretionary(XLY). This model finished last year with an 11% gain compared to a 26.4% gain for the broad market. Since 2001, this model has a CAGR of 4.1%, compared to -0.3% for the SPY. Hugh
Posted by hmTodd on Saturday, 01.30.10 @ 10:23am | #925
In addition to a moving average component, if the asset classes included cash since in some markets cash performs better then stocks (2008 for instance)I think the backtest models would then propel your strategy even farther ahead of the S&P.
Posted by Doug Nashif on Monday, 01.18.10 @ 18:10pm | #915
To answer a few questions:
Doug suggested the inclusion of something like a 200 day MA. I've done some testing on something along those lines and it's improved returns and risk. I want to write it up and post it when time is available.
Frank asked about using all sector funds. I've found the results are better if the funds are not too directly correlated. If you include all sector funds I think performance might suffer without some filtering mechanism.
Frank went on to ask about using this strategy with a fixed asset allocation. I'm not a registered investment adviser so I can't make a personal recommendation. That said, I take it that some people carve out part of their equity position to devote to this and/or other strategies. I've not done any backtesting on a bond specific strategy so I can't say with any certaintly how well that would work, but it sure makes sense.
Hugh
Posted by hmt on Monday, 01.11.10 @ 15:02pm | #893
Another question:
How do you integrate this system into an overall investment strategy? Does one keep a fixed asset allocation and select best ETF's within the asset classes? For instance in a 60%stocks/40%bonds does one sort the fixed income group and pick the best bond ETF.
Posted by Frank on Monday, 01.11.10 @ 07:29am | #892
Thanks for all the work. Have you looked at applying your technique to all the sector funds (All sector ETF's)? This technique would have one invested in bear market funds.
Posted by Frank on Friday, 01.8.10 @ 13:19pm | #890
Thanks for the great website and these two example strategy's. They look like excellent ways to invest part or all of one's portfolio especially if stop loss protection and a 200 dma moving average qualification are included.
Posted by Doug Nashif on Tuesday, 01.5.10 @ 05:22am | #889
December Update - This sector strategy gained 1.9% during the month of December, matching the return for SPY. For the year this strategy has returned 11.0% while the broad market gained 26.4%. This model currently holds Financials(XLF) and MidCap Value(IWS). XLF has dropped in ranking and will be replaced in this model with Consumer Discretionary(XLY) at today's close. Hugh
Posted by hmTodd on Monday, 01.4.10 @ 18:34pm | #886
November Update - This sector strategy gained 4.8% during the month of November while the broad market gained 6.2%. For the year this strategy has returned 8.9% while the SPY has returned 24.0%. This model currently holds Financials(XLF) and MidCap Value(IWS). These ETFs are still within the top screen rankings so there are no changes at this time.
Hugh
Posted by hmTodd on Tuesday, 12.1.09 @ 08:17am | #862
270_120.txt;8;12
Posted by fJyIUOIvTzI on Thursday, 11.19.09 @ 11:18am | #860
Oct Update - This sample strategy lost 5.7% during the month of October while the broad market, as measured by SPY, lost 1.9%. For the year this strategy has returned 4.0% while the SPY has returned 16.8%. This model currently holds Financials(XLF) and MidCap Value(IWS). These are currently in screen postitions 1 and 4, respectively, so there are not changes at this time.
Hugh
Posted by hmTodd on Monday, 11.2.09 @ 06:49am | #843
Todd - Prior to finding your site, I've played w/ several combinations of monthly averages. I'm still in XLF and XLB as I started with just the nine sector funds, but I now rank IWR and IWM for some sml/mid cap w/out overkill. Picking one point of 6 months scared me but we were never far apart until yesterday. For 6 months I use 24 weeks instead of 126 days. Comparing the 10/22 close vs. wk 5/4, 1)XLK 2)IWR 7)XLF, wk 4/27, 1)XLF 2)IWR, wk 5/11, 1)XLF 2)IWR. Average the 3 weeks and you still have 1)XLF 2)IWR. Scary! I'm wondering if you've ever back tested moving averages. When I average 126 and 200 MA, 1)XLF 2)IWR.
Great job, MIck
Posted by Mick on Friday, 10.23.09 @ 11:33am | #842
Sept Update - This sector strategy returned 5.3% in the month of September while the SPY returned 3.5%. For the year this strategy has returned 10.3% while the broad market has returned 19.1%. This model currently holds Financials(XLF) and Materials(XLB). Materials has dropped in screen rank since last month and will be replaced with MidCap Value(IWS) as of todays close.
Hugh
Posted by hmTodd on Thursday, 10.1.09 @ 08:43am | #839
Thanks for the heads up. The months were incorrect but have now been changed. As for the SPY return, I believe the 15% ytd number is correct although it doesn't feel like it. The screen rankings are taken as of the close of the month and XLB was in position #2 as of 8/31. On 9/1 it would have been IWS. Thanks for paying attention.
Hugh
Posted by hmTodd on Thursday, 09.3.09 @ 13:19pm | #837
Just a heads up I think you have made a mistake on your July and August updates. First you refer to the July update under your August update and then you list an update for June under your July update. Then you say that ytd the S&P 500 is up 15% through August. I don't think that is right. See below. Also how did XLB get into the top two? Was that as of 8/31 or 9/1. Thanks.
August Update - This sector strategy returned 2.5% in the month of July, underperforming the SPYs 3.7% return. For the year this strategy still trails with a 4.7% return compared to the broad markets 15.0%. This strategy currently holds Mid-Cap Growth(IWP) and Technology(XLK). These have both dropped in screen rank and will be replaced with Financials(XLF) and Materials(XLB) as of todays close.
Hugh
Posted by hmTodd on Tuesday, 09.1.09 @ 12:30pm | #834
July Update - This sector strategy returned 8.0% in the month of June, barely outpacing the broad market's 7.5% return. For the year this strategy is still seriously behind the SPY, with a 2.2% return compared to the market's 10.9%. This strategy currently holds Mid-Cap Growth(IWP) and Technology(XLK). These are still in the top six screen positions so there will be no changes this month.
Posted by d on Thursday, 09.3.09 @ 07:41am | #836
August Update - This sector strategy returned 2.5% in the month of August, underperforming the SPYs 3.7% return. For the year this strategy still trails with a 4.7% return compared to the broad markets 15.0%. This strategy currently holds Mid-Cap Growth(IWP) and Technology(XLK). These have both dropped in screen rank and will be replaced with Financials(XLF) and Materials(XLB) as of todays close.
Hugh
Posted by hmTodd on Tuesday, 09.1.09 @ 12:30pm | #834
July Update - This sector strategy returned 8.0% in the month of July, barely outpacing the broad market's 7.5% return. For the year this strategy is still seriously behind the SPY, with a 2.2% return compared to the market's 10.9%. This strategy currently holds Mid-Cap Growth(IWP) and Technology(XLK). These are still in the top six screen positions so there will be no changes this month.
Hugh
Posted by hmTodd on Saturday, 08.1.09 @ 12:28pm | #831
For anyone reading, Dean and I have been conversing about his market timing tweak via email and hope to share some results soon.
Hugh
Posted by hmTodd on Saturday, 08.1.09 @ 12:24pm | #830
Hugh:
I would like to see how your model would be impacted by the addition of a simple market timing element to the mix. For example, stay in the ETF's only when the S&P 500 index is above its 200 day moving average, while moving to cash anytime this index falls below its 200 day moving average. My guess is that this small tweak would actually improve the performance of the model.
Dean
Posted by Dean on Thursday, 07.30.09 @ 15:35pm | #829
Cheryl,
The ytd return through June was -5.3%. Sorry I failed to include it in the update.
I wish I could help you on how to select the strategy and how to implement it. But since I am not a licensed investment advisor I should limit myself to providing data and letting you determine if and how it might fit your needs. That said, note these sample strategies are aggressive and historically have had significant losses at times. And remember, past performance is no guarantee of future performance.
Hugh
Posted by hmTodd on Thursday, 07.30.09 @ 12:58pm | #828
For the June update, you forgot to mention the YTD performance of the sector strategy vs the SPY. Please be sure to give this info in the July update. Thanks. I am trying to decide which strategy to use and this would be helpful.
Posted by Cheryl on Thursday, 07.30.09 @ 12:24pm | #827
I just found your website and am interested in implementing your sector strategy. Should I start with the top two as of today's date, which are XLF and XLK, or should I start with the two that were on the top back on July 1st, which were____ and ___?
Posted by Cheryl on Tuesday, 07.21.09 @ 14:35pm | #826
Can you create and test a strategy for inverse ETFs when the broad market indices trend is down? It would be interesting to use this long sector strategy when the broad markets are up and then sell all long positions and buy inverse positions when the markets are trending down. Just a thought.
Posted by chris on Friday, 07.17.09 @ 18:57pm | #825
June Update - This model strategy returned 0.4% in the month of June while the market (SPY) returned -0.1%. This strategy began the month holding Consumer Discretionary(XLY) and Mid-Cap Growth(IWP). XLY has fallen out of the top 6 screen positions so was replaced by Technology(XLK) as of the close on July 1st.
Hugh
Posted by hmTodd on Monday, 07.6.09 @ 09:57am | #822
Yes this page should be updated and I apologize for not getting to it sooner. I'll update the model results in another post but get to the recent questions below.
First for the 2007 results. The correct performance is 11.3%. I initially reported a 3.4% loss for July but the correct number was -1.3%. Sorry about that, and a good catch on your part.
Have we tested RSF as opposed to TR126? Yes. In fact I just ran the numbers again and TR126 won for the period 2001 - current, by about 0.5% per year. This is somewhat surprising to me as I would have expected RSF to perform better as it seems to on most groups.
Have we tested daily returns? Yes. I don't have those numbers in front of me at the moment, but the increased trading generally results in selling short turn sell-offs and buying short term blips that don't hold up long enough to profit from the trade. My experience is that you need some mechanism to prevent this and with these models we use time for simplicity.
Again, sorry about the delayed responses.
Hugh
Posted by hmTodd on Monday, 07.6.09 @ 09:50am | #821
Is this page still being updated?
Posted by bruce on Monday, 07.6.09 @ 08:11am | #820
have you ever tested a model using your existing rules, but running the screen daily? I was wondering if it improved performance to continuously monitor this list and then take action whenever a positon drops below your established threshold.
thanks for your time- you have a great site
Posted by bruce on Tuesday, 06.23.09 @ 07:11am | #819
Have you tested your own relative strength measure, RSf, in addition to using 6 month performance? If not, why not? If so, how did it do compared to the 126 day measure of relative strength? Thank you. This is a great site and an excellent resource.
Posted by Malcolm Trevillian on Friday, 06.5.09 @ 11:00am | #811
I notice that in January of '08 you list your 2007 performance as +9.2 % but in August of '08 you show 2007 as +11.3%. What accounts for the difference?
thanks
Posted by bruce on Thursday, 06.4.09 @ 10:23am | #810
May Update - The SPY gained 5.8% this month and moved into positive year-to-date performance with a gain of 3.3%. This model strategy gained 2.3% for the month and still has a loss of 5.7% for the year. The current holdings, Consumer Discretionary(XLY) and Mid-Cap Growth(IWP), are currently in screen positions 4 and 1, respectively, so there are no changes this month.
Hugh
Posted by hmTodd on Monday, 06.1.09 @ 07:38am | #808
I scanned the previous letters and suspect some of my comments may have been answered.
Have you experimented with a modification of the "rules" ?
What if you used a 3 or 1 month performance criteria instead of a 6 month period?
What if you traded and bought the top two funds each month with no consideration of previous rank?
I realize that trading costs will increase but this is minimal these days. Slippage would be a factor, and I'm not sure of the overall advantages. I thought you might have done the research. Great site. Thanks for the help.
mike
Posted by Radman on Sunday, 05.17.09 @ 12:04pm | #807
this post is fantastic
Posted by Zfdtffdo on Sunday, 05.10.09 @ 17:11pm | #806
Wonderfull great site
Posted by Pilzdgwp on Sunday, 05.3.09 @ 00:38am | #805
I'm happy very good site
Posted by Ssjomxuc on Friday, 05.1.09 @ 23:44pm | #804
April Update - The broad market continued its positive move in March, gaining 9.9% for the month and cutting its year to date loss to 2.4%. This strategy gained only 2% during the month and currently has a loss of 7.8% for the year. The two positions in this model portfolio have been Consumer Staples(XLP) and Healthcare(XLV). These defensive holdings have given way to Consumer Discretionary(XLY) and Mid-Cap Growth(IWP) as of todays close.
Hugh
Posted by hmTodd on Friday, 05.1.09 @ 10:29am | #802
Best Site good looking
Posted by Atiakqdj on Friday, 04.24.09 @ 21:51pm | #801
March Update - The market had a nice bounce in March. The two positions in this strategy returned 5.2% in the month while the SPY returned 8.3%. For the year the market has now lost 11.2% while this strategy has lost 9.6%. XLV and XLP remain in the top 6 screen positions so there are no changes to the model this month.
Hugh
Posted by hmTodd on Wednesday, 04.1.09 @ 07:48am | #799
Shawn asked about an update similar to the 8/8/2008 update. The 2001 - 2007 numbers obviously did not change. Below are the numbers for 2008 to present.
Year Sector Int'l SPY
Strategy Strategy
2008 -44.4 -46.8 -36.8
2009 -9.6 -20.2 -11.2
CAGR 2.2 5.3 -4.1
2008 hurt the long term numbers but the two strategies still remain above the SPY.
Hugh
Posted by hmTodd on Wednesday, 04.1.09 @ 07:44am | #798
can you provide some update as to the overall results since starting the strategy as you did previously on 8/8/2008 (#734)
thanks
Posted by Shawn on Monday, 03.30.09 @ 13:42pm | #797
Look at the UYG at the 52 week low and ready for a huge run. The chart is looking greater every day load up
Posted by JJ on Monday, 03.9.09 @ 17:07pm | #796
February Update - As I write this the headlines are about the Dow losing half its value since its 2007 peak. That says a lot about the market as a whole. This momentum strategy lost another 10.4% during the month while the SPY lost 10.7%. For the year this strategy has lost 14.1% while the broad market lost 18.1%. The two holdings of this model, Consumer Staples(XLP) and Healthcare(XLV), remain in the top two screen positions so there are no changes at this time.
Hugh
Posted by hmTodd on Monday, 03.2.09 @ 08:21am | #792
January Update - The two holdings of this model, Consumer Staples(XLP) and Healthcare(XLV), lost a little ground in January, with the model portfolio losing 4.1% while the broad market, as measured by SPY, lost 8.2%. These are still in the top two positions so there are no changes at this time.
Hugh
Posted by hmTodd on Monday, 02.2.09 @ 12:39pm | #788
This remains one of the best ETF sites available. Thanks for maintaining it and for posting monthly performance results. I find your ETF Performance screen especially helpful. I use it currently in the 3 month setting.
Posted by kayboe on Sunday, 01.18.09 @ 18:51pm | #787
December Update - The two holdings of this model, Consumer Staples(XLP) and Healthcare(XLV), performed well in December returning 5% and 12%, respectively, from 12/1 to 12/31. But the strategy lost 3.0% because of the losses in IWO and IWM on the first day of December. The broad market(SPY) returned +1.0% over the same period. For the year, this strategy lost 44.4% while the SPY lost 36.8%. With that, 2008 is done. The holdings, XLP and XLV, are still in the top two screen positions so remain in this strategy for January.
Hugh
Posted by hmTodd on Friday, 01.2.09 @ 06:25am | #785
November Update - The market turmoil continues and the performance of this strategy continues to reflect that environment, losing 11.9% last month while the SPY lost 7.2%. For the year this strategy has lost 42.7% while the broad market has lost 37.6%. This strategy began the month holding IWO and IWM. Both of these have now dropped in rank and will be replaced by Consumer Staples(XLP) and Healthcare(XLV).
Hugh
Posted by hmTodd on Monday, 12.1.08 @ 08:13am | #782
October Update - The market turmoil took its toll on this model portfolio this month with a loss of 21.3% as compared to a market loss of 16.5%. For the year to date, this strategy has lost 34.9% while the SPY has lost 32.7%. Both IWO and IWM are still in the top 6 screen positions so there will be no changes this month.
Hugh
Posted by hmTodd on Monday, 11.3.08 @ 09:12am | #779
First, thanks again for maintaining this site.
Have you tested using RSf with this strategy? (I am sure you have, interesting results.)
I would love to see this screener enhanced to allow:
*a user chosen pool of etfs.
*a hold top #.
*and while I am at it, a HTD feature.
I scanned, and see similar questions were asked in the past, I hope you don't find these questions/requests annoying.
thanks again,
Posted by Editedby on Sunday, 10.26.08 @ 14:11pm | #778
Posted by on Friday, 10.24.08 @ 07:15am | #777
Posted by order viagra Text=order viagra online order viagra on Sunday, 10.19.08 @ 20:42pm | #776
I tried to use your ETF screener to replicate your ETF strategy and had trouble selecting the right criteria.
Could you please tell me what criteria you would enter into the screener to obtain the rankings for the sector ETF strategy?
Thank you.
Posted by Investor237 on Saturday, 10.4.08 @ 11:53am | #775
September Update - The two Russell funds in this model portfolio offered no protection from the market's correction. This portfolio lost 13.1% during the month while the broad market lost 9.4% as measured by the SPY. For the year this strategy is down 17.3% while the SPY is down 19.4%. Since IWO and IWM are still in the top 6 screen positions there will be no chances this month.
Hugh
Posted by hmTodd on Wednesday, 10.1.08 @ 18:43pm | #773
Good screener.
Have you never tried how a different review time and/or picked ETFs could effect performance and risk ?
Probabily a shorter time (maybe 1 week) and a reduced number of ETF would increase risk (and costs) but it will be interesting seeing effect on reward (Sharpe)
Posted by Stefano on Monday, 09.22.08 @ 13:58pm | #772
August Update - August continued the shift away from commodities and into equities, although at a cautious pace. We began the month holding Materials(XLB) end Energy(XLE) and both lost ground during the month. During August this model strategy lost 0.5% while the market returned +1.5%. For the year they have lost 4.8% and 11.0%, respectively. The model is indicating it is time to close the two open positions as of today's close and move into two Russell funds, Small Cap Growth(IWO) and Small Cap in general(IWM). Interestingly, the last time these two funds were held at the same time was in 2003.
Hugh
Posted by hmTodd on Tuesday, 09.2.08 @ 11:08am | #769
I found your strategy very interesting! I do have some questions though:
1. What is the criteria for you selections?
2. How is the ranking determined?
Posted by Thom on Sunday, 08.17.08 @ 06:30am | #767
Paul,
Always interesting to see how different researchers can come up with such different results. Since I always like to learn I would appreciate a copy of your study so I could review it in more detail. My email address is at the bottom of the home page.
I'm glad you reminded our readers about transaction costs as these can quickly erode returns with systems that trade often. That is one of the advantages of this system, it has an average holding period of over six months so transaction costs are much less significant. Slippage costs are also closely tied to liquidity with the more liquid ETFs having very tight spreads. As I write this the ETFs in this strategy are typically trading at a 1 cent bid-ask spread with a 2 cent spread showing up occasionally. With the less liquid ETFs spread can difinitely become a factor so I'm glad you mentioned it and it points out another advantage to trading the ETFs in this strategy.
Calculating annual losses due to slippage where slippage is 1 cent on a $30 stock I get a 0.13% loss annually figuring a six month hold. Lets basically double it and call it a quarter of a percent. There are other costs but Im still not seeing 5% or 2x 5% if the 'bid-ask gap' is the largest component. But I definitely could be wrong.
Looking forward to a copy of your study so I can better understand.
Sincerely,
Hugh
Posted by hmTodd on Monday, 08.11.08 @ 12:36pm | #749
A new, carefully performed study by two Dutch researchers concludes that ETF's cannot be used to profitably trade U.S. sectors using a momentum strategy. They found a mean 5% return per year without any trading costs, but theier estimate of the trading costs --- which was quite optimistic and which was less than those costs imputed in the ETFScreen scheme were more than 5% per year.
We at P & B Investments have performed more comprehensive studies involving Ishare data from 1996 to 2008 and confirmed their results. The net return from a sector-momentum strategy is a very sensitive function of trading costs. We believe the "real-life" trading costs (largest component is bid-ask gap or "slippage") are underestimated by the Dutch researchers and could be as much as 2X their estimated costs for the back-test period.
Our studies examined a large family of possible momentum stratgies, including the ETF Screen strategy --- which is one of the better ones.
Summary: The arbitrage represented by any ETF sector momentum strategy is completely "traded-away" by the Market-Makers or Principals in the ETF markets. Small investors can only achieve stochastically insignifcant gains from trading a momentum ETF strategy.
Paul H. Lasky
P & B Invesments.
Posted by Paul H. Lasky on Monday, 08.11.08 @ 08:50am | #748
Some wanted an update on how this strategy has done longer term so here is the data. The period covers 2001 through July, 2008. This is just an extension of the period covered above in the initial release.
Posted by hmTodd on Friday, 08.8.08 @ 08:24am | #734
July Update - Due to an unexpected trip this is going to be short and to the point. I'll try to post more and update return data next week. This model strategy lost 11% during July when the SPY lost 0.9%, bringing the year to date returns to -4.4% for this strategy and -12.4% for the broad market. We've been discussing when the run for Energy and Materials would be over and it might have begun. But these funds are still in positions 1 and 3 so they will stay in the portfolio for another month.
Hugh
Posted by hmTodd on Thursday, 07.31.08 @ 19:44pm | #712
GreggB - I'd like to find a screen utilizing the short ETFs that are now available, but haven't devoted time to it recently. One problem in backtesting the short ETFs is their lack of price history. One would probably need to start by synthesizing index values for a longer history. Thanks for the question. Hugh
Posted by hmTodd on Tuesday, 07.29.08 @ 06:35am | #703
Sorry. Didn't see Adam's similar short question from 2006.
Posted by Gregg B on Monday, 07.28.08 @ 14:00pm | #698
I have not started to use this strategy yet but may begin the first of August.
any thoughts on doing a similar strategy for Short ETF's? To take advantage of segment downturns as well as segment momentum? Just a thought.
Posted by Gregg B on Monday, 07.28.08 @ 13:55pm | #697
June Update - June was a miserable month for the market but this strategy came through it with only minor damage. For the month this model portfolio lost 0.5% while the broad market lost 8.4%, taking year to date numbers to +7.5% for this model and -11.6% for the SPY. According to my numbers this was the worst monthly performance for the SPY since Sept 2002, when the world was expecting another terrorist attack.
Last month I wrote about not knowing when the current trends in Energy and Materials would end and we still don't. But they weren't bad places to be this past month and we'll continue holding them until something else rises to the top of our screen.
Hugh
Posted by hmTodd on Tuesday, 07.1.08 @ 07:31am | #692
Michael, You could probably have improved performance the past year with some other commodity funds in the mix. Of course, that is hindsight at this point in the cycle. Part of the theory behind this model is to limit the selections to funds with low correlation so the family of Select Sector SPDR's was used.
The exit rule used allows funds room to move without exiting in the middle of a move - at least that is the goal. If your exit is too tight you risk exiting on a minor pullback and then buying again at a higher price. Of course nothing is exact in this business.
Thanks for the comments,
Hugh
Posted by hmTodd on Tuesday, 07.1.08 @ 07:22am | #691
Hugh,
Great stuff here. I was wondering if you ever thought about adding some of the commodity funds out there? Granted the XLe and XLB will catch most of the companies that do commodities.
Additionally, Have you considered changing the "exit" rules? I was wondering what an exit rule of below position 3 would look like?
Posted by Michael on Thursday, 06.26.08 @ 04:58am | #690
Thanks Hugh, I look forward to your posts every month.
Posted by editedby on Monday, 06.16.08 @ 09:06am | #685
May Update - May was another good month for this model portfolio, returning 5.2% while the broad market returned 1.5%. For the year this strategy has returned 8.0% and the SPY is still negative at -3.5%. You never know when the market will turn against the energy and materials sectors but it will some day. For the moment we will stay where we are holding XLE and XLB because they are top in the screen.
Mike - We're still here, just a little late getting the monthly update posted. There have been no tweaks since the original concept. The updates have been in the comments section so far but need to be reflected in the general information at some point - it is a little dated.
Hugh
Posted by hmTodd on Monday, 06.2.08 @ 07:51am | #672
Has this site closed down? It is now June 2008--
Have you continued to collect data and tweak the system.
Any current impressions.
Posted by mike on Saturday, 05.31.08 @ 09:08am | #670
April Update - The month of April provided a good snap back for the market and this strategy capitalized on that, returning 8.3% while the broad market returned 4.8%. For the year this strategy is now in the black at +2.7% while the SPY is still negative at -5.0%. XLE is still #1 in the rankings and XLB is #3 so there are no portfolio changes this month. Hugh
Posted by hmTodd on Thursday, 05.1.08 @ 05:44am | #661
March Update - The tough market continues. This strategy lost 1.9% over the month holding Materials(XLB) and Energy(XLE) while the SPY lost 0.9%. For the year this model portfolio is ahead, if you can call it that, losing 5.1% while the market has lost 9.3%. Neither are very good. Going forward, the above two holdings are still in positions 3 and 2, respectively, in our list so they will remain in the portfolio another month. Hugh
Posted by hmTodd on Tuesday, 04.1.08 @ 07:25am | #647
HZ -
I don't have the system and data needed to perform such a test, sorry. It would be interesting to see the results if you put together a backtest. In putting this strategy together initially I drew from my experiences in stock trading systems and these momentum type screens seem to be fairly good long term performers. But they are not perfect. - Hugh
Posted by hmTodd on Thursday, 03.6.08 @ 06:34am | #636
Hugh,
Is it possible for you to extend your testing period? For example using some fund mutuals that exists starting 1990? From 1990 to now, we see a complete market cycle - booming, bubble, bubble burst and now (a dead market). It will be more convincing if your strategy works in different market condition.
HZ
Posted by HZ on Wednesday, 03.5.08 @ 14:12pm | #635
HZ -
I haven't tested your S&P 500 strategy, but my experience with market timing strategies that are either in the market or out of the market shows they need more smarts than a simple trailing return. This strategy does not try to time when to be in and when to be out, it merely tries to stay in the top performing segment. One could probably improve on this return with some rules that would exit the market entirely under certain circumstances, but that is more complex than I wanted to get with this simple strategy. Maybe something I should look at as a more advanced strategy.
Thanks for your comments,
Hugh
Posted by hmTodd on Wednesday, 03.5.08 @ 13:46pm | #634
Hi I found your website when searching ETF strategy on google. Your provide some very interesting results. However, since ETFs are relatively new, it is difficult to test your strategy across different market condition.
I use a similar though simple momentum strategy to test SP500 since I can get the data back to 1950 on yahoo’s website. My strategy is starting from 1 share of “SP500” then every month if the SP500’s return (I used 1, 3, 6 and 12 months return) is below 0, I sell it, otherwise I hold it. After selling of “SP500”, if the return becomes positive, I purchase it back.
The strategy actually works really poor from 1950 till yesterday. Basically if I start with $17 in 1950, buy and hold till yesterday will give me $1326. However, the above strategy will give me $332 (use 1 month return), $307 (3 months return), $647 (6 months return) and $878 (12 months return).
Interestingly, if I change the testing period to 2000.1 till yesterday. The strategy works well, it gives me $1507 (1 month return), $1447 (three mos), $2185 (six mos) and $1830 (12 mos) comparing with $1326 using buy and hold.
Posted by HZ on Wednesday, 03.5.08 @ 13:19pm | #633
February Update - It was another unpleasant month for the market, and one that looked at times like it could be so good. This sector strategy held the two best funds by a long shot, returning +6.2% during the month when the SPY lost 2.6%. That still wasn't enough to carry us into positive territory for the year but it was a start. For the year this model strategy has lost 3.3% while the broad market has lost 8.5%. Our holdings, XLE & XLB, are at the top of the rankings so there are no portfolio changes at this time. Hugh
Posted by hmTodd on Saturday, 03.1.08 @ 07:39am | #631
It looks like I missed a couple of questions being asked. First for the question about combining returns from various periods, like 3 months, 6 months, and 12 months. Such a model is used by some and is effective. There is no one right way and many that will keep you in the top tier of performance over the long haul.
As for Werner's question about returns not matching. I would have expected Yahoo's results to match the numbers I use. In fact I have used their numbers to verify my results at times past. The difference is probably in the number of days used. For monthly increments I use a consistent 21 market days per month. That removes some of the anomolies other methods experience. If you did not use 126 days for the 6 month period you might look at it again.
Sorry for the delayed reply,
Hugh
Posted by hmTodd on Saturday, 03.1.08 @ 07:31am | #630
I have read your very interesting work about ETF-Screening.
I tried to write a rotationsmodel for my own and took your idea
as an inital point.
I was using quotes from yahoo finance. But also with yahoos
adjusted Prices I could not replicate your entries and exits.
Could you please tell me which prices you where using
of which type of adjusting?
Best Regards,
Werner Dunker
Posted by Werner Dunker on Sunday, 02.17.08 @ 05:39am | #624
What about using a simple combination of returns: add the 3 month, 6 month, and one year returns and rank based on that sum. Or, to emphasize recent performance: add 1 month, 3 month, and 6 month returns.
This additive system is one of the variations used to rank mutual funds in Gerald Appel's book Opportunity Investing. (He created the MACD indicator.)
I would love to see columns in your return tables that would sum the returns so we could sort them. Difficulty is in deciding which returns to add.
Thanks for a great site!
Posted by uthr on Friday, 02.15.08 @ 14:41pm | #623
Phil asked a good question about using the one month trailing return instead of the six month. From my experience and tests I've performed in the past I've found a short term measurement results in many more trades and whipsaws that eat into returns. In short, you react to the noise rather than changes in trend. Sorry I don't have any numbers on hand at the moment to support this.
Posted by hmTodd on Thursday, 02.14.08 @ 11:58am | #622
what about using a one month trailibg return instead of six month?
Posted by Phil Kilby on Thursday, 02.14.08 @ 09:23am | #621
January Update - No hiding from the down market with this strategy which got hit like everything else did. We began the year holding positions in Materials(XLB) and Energy(XLE). This combination lost 8.9% for the month, while the SPY lost 6.0%. Not the way we like to begin but the year has a lot left in it. These two funds are still among the top 6 on the list so there are no portfolio changes at this time. Hugh
Posted by hmTodd on Friday, 02.1.08 @ 06:20am | #615
December Update - This model portfolio began the month holding positions in Materials(XLB) and Energy(XLE). XLE had a strong month and this strategy returned 5.1% in the month while the broad market lost 1.1%. For the entire year of 2007 this portfolio returned 9.2% while the SPY returned 5.1%. These two funds are still in the top 6 screen positions so there are no changes this month. Wishing each of you a Happy and Prosperous New Year. Hugh
Posted by hmTodd on Tuesday, 01.1.08 @ 09:28am | #609
November Update - This model portfolio began the month holding positions in Materials(XLB) and Energy(XLE), which resulted in a loss of 4.6% for the month compared to a 3.9% loss for the SPY. For the year this strategy has returned 3.9% while the market, as measured by the SPY, has returned 6.3%. These funds are currently ranked #5 and #1 on our screen so there are no changes again this month. Hugh
Posted by hmTodd on Monday, 12.3.07 @ 05:50am | #598
There are any number of successful approaches to trading ETFs. The strategies presented here as examples are mainly to generate ideas and show what can be done. From the backtesting I have perfomed you don't want to switch from one fund to another if the two are highly correlated. That is one of the reasons the strategies on the site use a fixed list of funds, trying to avoid overlap. I see from the comments that you all recognize this and are avoiding the related funds. Stops are a difficult issue. I have seen few cases where a percentage stop improves returns for a momentum based strategy. Psychologically, however, stops make a trading strategy more palatable and can help to avoid some big losses in exchange for more small ones. That said, I have not tested stops on these ETF strategies so cannot comment with quantitative support.
Good luck to each of you and I am glad to see the exchange of ideas.
Hugh
Posted by hmTodd on Monday, 12.3.07 @ 05:44am | #597
Becca I'm doing a similar thing with the "sector" strategy. I'm taking Hugh's list from the menu on the left of "All Sector ETF's". Sorting it by 6 months and looking for the best unrelated ETF's. I'm also sticking mostly with the iShares and SPDR because the other flavors seem to tend to putting a large percentage in a single stock which I don't like. I'm not using any stop though other than following Hugh's methodology of trading out any that have fallen out of the 8th position (since it is a long list).
Posted by Eric on Friday, 11.30.07 @ 11:29am | #596
Re: Becca's strategy
Many professional managers have used simple momentum strategy successfully like your strategy. Two name comes to mind are: fundxfunds.com and ETF and sector momentum tracker from fidelityadviser.com. Also I beleive etfquest.com uses RS based strategy.
Posted by investor on Thursday, 11.29.07 @ 11:38am | #595
I am thinking of using a strategy very similar to that posted by Becca on 10/3/07 in comment 854, both in terms of ETF selection criteria (although mine are slightly different) as well as the use of a stop loss (I am using a 10% trail stop).
I was wondering if you had any feedback on this type of strategy, and I am also wondering what you think of including the leveraged ETFs such as QID (i.e., 2 x QQQQ) in your system.
thx Mike
Posted by Mike on Sunday, 11.11.07 @ 21:55pm | #594
October Update - This portfolio began the month holding positions in Materials(XLB) and Energy(XLE). These holdings yielded a 3.2% gain for the month taking the year-to-date return to 8.9%. This compares to a 1.4% monthly gain, and 10.6% ytd gain, for the SPY. These funds currently rank #3 and #1 on our screen so there are no changes this month. Hugh
Posted by hmTodd on Thursday, 11.1.07 @ 11:16am | #592
I'd like to know what you think about the following strategy using your performance screen of ALL etfs (Last time I counted I think there were about 240!) What I'd like to do is buy the top three UNRELATED etfs at the eginning of the month (for example I would not buy two China funds, nor a China fund and a BRIC fund) based on 6 month return. It would not matter to me whether they fall into international or sector strategies or both. Whatever etfs are in the top 15 percent of all etfs listed at the beginning of the next month would be kept, while those falling below this cutoff would be sold and replaced. Another thing I would add is a stop-loss that would automatically sell etfs that lose more than 10 percent during the month. When this happens I would wait until the first of the following month to buy an etf to replace the stopped out position, even if that means re-entering that stopped out position! I would commit 20 percent of my retirement funds to this sector strategy while the rest of my funds would be in diversified stock and bond mutual funds. I would use Morningstar to determine whether the top etfs are sufficiently different to own concurrently. Thanks so much for your web site and thanks ahead of time for your opinion of my plan.
Posted by Becca on Wednesday, 10.3.07 @ 10:50am | #584
September Update - This portfolio began the month holding Materials(XLB) and Energy(XLE). Luckily these were the two top performing sectors for the month and this model portfolio returned 7.8%, taking the year-to-date return to 5.5%. The SPY, for comparison, has returned 3.9% for the month and 9.1% for the year. Obviously we are still behind for the year and need a strong finish to catch up. Since both of these funds are still in the top 6 positions there will be no changes at this time. Hugh
Posted by hmTodd on Monday, 10.1.07 @ 04:49am | #581
August Update - This model portfolio held Materials (XLB) and Utilities (XLU) during the month of August. The net result was a gain of 0.6% for the month taking the year to date return to -2.1%. For reference the SPY gained 1.3% in August and 5.1% year to date. XLU has dropped in rank and will be replaced with Energy (XLE) in this model portfolio on the close today. XLB is currently in position #6 which will keep it in the portfolio for another month.
Hugh
Posted by hmTodd on Tuesday, 09.4.07 @ 06:02am | #570
July Update - This will be a two month update since there was no update posted in June. This model strategy lost 2.9% in June and another 3.4% in July yielding a ytd return of -2.7%. The comparable numbers for the SPY were -1.5 and -3.1 for June and July and +3.7 year-to-date. This strategy is still holding materials (XLB) and utilities (XLU). We'll see where they go in August.
Hugh
Posted by hmTodd on Wednesday, 08.1.07 @ 11:54am | #564
The normal monthly update will have to wait a week or so, but I took a quick glance at market close today and it looks like both XLB and XLU will remain in this model portfolio another month even though XLU had poor performance in June.
Posted by hmTodd on Friday, 06.29.07 @ 19:02pm | #480
Update for May - This sector portfolio returned 2.7% in May holding Materials(XLB) and Utilities(XLU), taking the year to date return up to 3.8%. While positive, both underperform the SPY which returned 3.4% in the month and 8.7% y.t.d. Since both of the above funds are still near the top of our list they remain in the model portfolio for another month.
Posted by hmTodd on Friday, 06.1.07 @ 09:20am | #453
Update for April - This model portfolio returned 2.3% in April holding Materials(XLB) and Utilities(XLU), taking the year to date return to a positive 1.1%. For comparison, the SPY gained 4.4% for the month and 5.1% for the year. This strategy continues to lag the market, but has now turned positive on the year. We'll see how May goes with these same two funds since they are still at the top of the screen list. In other words - no changes in holdings this month.
Posted by hmTodd on Tuesday, 05.1.07 @ 05:49am | #421
Update for March - Our sector stratgy portfolio returned a positive 0.6% in March while the market, as measured by SPY, returned 1.2%. For the year this model portfolio is down 1.4% compared to a positive 0.7% for the market. Our Consumer Discretionary(XLY) holding has fallen in our screen during March and should be replaced by Utilities(XLU) as of today's close.
Posted by hmTodd on Monday, 04.2.07 @ 09:05am | #400
Josh asked about monthly performance numbers - Yes the numbers used on this site take dividends into account since they are returns just as capital gains are. Another reason we could be different is that we always use the same number of trade days for a given period where many sites use a calendar. We consider a month to always have 21 trading days so we effectively remove one variable from interpreting the results.
Posted by hmTodd on Monday, 04.2.07 @ 09:00am | #399
How do you determine the monthly returns? They do not seem to match up with what I see elsewhere on the web. It looks as if you take into account dividends but I don't think that should make that much of a differnce (for example: according to Ishares website the 3 month performance for IWP is 2.43% , you have it as 4.01% , a fairy big difference, just wanted to get an idea how this is figured)
Thanks
Posted by Josh on Wednesday, 03.28.07 @ 06:22am | #396
Great site and a very interesting approach to ETF investing. I noticed that the most current data only runs through Sept 05 and I wanted to ask you when (or if) you were planning on posting results through 2006.
Thanks in Advance,
Robert
Posted by Aquafiend on Sunday, 03.4.07 @ 20:27pm | #327
Update for February - After Tuesday's market drop this strategy lost 2.6% in the month of February, bringing the ytd return down to -2.0%. Technology(XLK) has not performed well and has dropped out of the top 6 screen positions so it will be removed from this model portfolio on today's close with the proceeds going into Materials (XLB).
Posted by hmTodd on Thursday, 03.1.07 @ 11:49am | #282
Great site you have - I just stubled across it. I have two questions.
What exactly is RSf, and is it a standard technical indicator.
When you calculate return, do you include the dividends? Thanks.
EARL
Posted by Earl on Sunday, 02.25.07 @ 17:14pm | #280
An excellent and simple strategy.
Do you have any plans to post the 21 funds current position or is this just an historical survey?
Posted by cma on Saturday, 02.17.07 @ 18:59pm | #276
I believe No Load Fund X newsletter out of San Francisco is ranked number one over 15 years by Hulbert and they pretty much have the same format. They pick the top couple ranked mutual funds and as they drop out of the top 5 or so then they select new ones. So the basic format has been successful for many years.
Of more recent note the Coolcat newsletter for ETFs has a similar strategy with a couple qualifiers and has been top ranked by Hulbert in recent years.
Have yu thought about putting up a section that just uses your strategy.
Simply a great site. I especially like that you have adopted heatmap color coding concept with your spreadsheet listing.
Posted by rw mac on Saturday, 02.10.07 @ 19:20pm | #275
Wow!!!
I just stumbled accross your sight today. You should be commended for the time, energy you have put into this sight. You are in "my favorites". Would you consider any inverse ETFs in your mix? Have you backtested any RSf strategies? Keep up the good work, it is appreciated.
Ralph
Posted by Ralph Regula on Friday, 02.2.07 @ 11:23am | #272
Update for January - This Select Sector SPDR strategy returned 0.6% in January while the market as measured by the SPY returned 1.5%. Our Utilities(XLU) position lost a little while our position in Technology(XLK) picked up a couple of percent during the month. XLU has now dropped in the rankings and will be replaced in this model portfolio with Consumer Discretionary(XLY) on today's close.
Posted by hmTodd on Thursday, 02.1.07 @ 07:47am | #269
Update for December - December completed a positive but sub-par year for this model portfolio. During the month this model lost 0.3% while the SPY gained 0.8%. For the year the model returned 8.3%, respectable but just over half of the SPY's 15.2%. Since the beginning of 2001 the model returned 13.6% versus 3.1% for the broad market. The portfolio holdings are Utilities(XLU) and Technology(XLK) and do not change this month since both positions are still in the top 6 screen positions. Have a happy New Year and may your 2007 be a great year.
Posted by hmTodd on Sunday, 12.31.06 @ 10:58am | #262
Update for November - This model portfolio returned 1.2% during the month holding Utilities(XLU) and Consumer Staples(XLP). The market during this period gained 2.0% as measured by the SPY. For the year this strategy has gained 8.5% while the SPY has gained 14.3%. The Consumer Staples(XLP) sector lost ground during the month and finished near the bottom of our list of funds. That position should be closed as of today's close and replaced with Technology(XLK).
Posted by hmTodd on Friday, 12.1.06 @ 05:59am | #250
Update for October - This model portfolio held Utilities(XLU) and Consumer Staples(XLP) since October 1st, and returned 3.6% for the month compared to the SPY's 3.2%. This brings our ytd results to 7.2% during which time the market has gained 12.1%. The Utilities fund is still the top fund on our list and Consumer Staples is in position #4 so there are no portfolio changes.
Posted by hmTodd on Wednesday, 11.1.06 @ 06:35am | #113
Hello again,
Just another question. How do derive the numbers
for your Relative Strength Trends (RSf) charts,
both the weekly and monthly versions.
Thanks, William Thomas
Posted by William F Thomas on Monday, 10.23.06 @ 14:23pm | #70
Hello,
I just have a couple of questions.
It looks like in your sector strategy (and perhaps also the foreign etf stategy)you are
always in the market, never on the sidelines.
Would that not be a problem in a world wide
bear market? Is there a way of moving to the
sidelines during such an episode using these
strategies?
Thanks, William Thomas
Posted by William F Thomas on Friday, 10.20.06 @ 21:41pm | #69
Update for September - We began the month holding Utilities(XLU) and Energy(XLE). For a second consecutive month the market was not kind to the energy fund and our model portfolio lost both in absolute terms (-2.4%) and against the market, which gained 2.7% on the month. This brings our ytd results down to 3.5% during which time the SPY has gained 8.7%. The model indicates its time to close out the XLE trade and shift those assets into Consumer Staples(XLP) as of today's close.
Posted by hmTodd on Monday, 10.2.06 @ 05:19am | #67
Update for August - We began the month holding Utilities(XLU) and Energy(XLE). The month was not kind to the energy fund and our model portfolio lost 2.2%. This brings us down to 6.1% for the year compared to the market's 5.8%. The SPY gained 2.2% in the month of August. Since both funds are still in the top 6 positions (XLE is #6) there will be no trades this month.
Posted by hmTodd on Friday, 09.1.06 @ 06:14am | #64
Geoff,
Three month trends may be tradeable with a similar model but the trades will be more frequent. I do not have any numbers in front of me at the moment to judge relative performance, but with more trading comes increased trading costs.
Hugh
Posted by hmTodd on Friday, 09.1.06 @ 06:10am | #63
It seems that a 3 month performance period might be more prudent than the 6 month, since most sectors and styles are generally peaking within 6 months. With the 6 month selection you will be buying a sector near the top of it's cycle.
Is this wrong?
Posted by geoff on Saturday, 08.26.06 @ 06:26am | #62
Update for July - This sector based strategy had a small loss in July as the small cap value fund lost more than the energy fund gained. For the month the combination lost 0.8%, more than the markets loss of 0.4%. For the year this strategy has returned a theoretical 6.5%, beating the SPY's 2.6%. The trading rules call to sell IWN at the end of today's trading and buy utilities (XLU) with the proceeds.
Posted by hmTodd on Tuesday, 08.1.06 @ 08:40am | #59
This is the most useful site for ETF that I have seen so far.
One feature to the screener that could value: (this is the one i sometimes use in stock selections)
"Ability to pick TOP n ETF for a criteria and be able to do more than one time"
(currently i see we can only do one level sort)
One I would like to do personally is something like -
TR126 top 20
TR63 top 10
TR21 top 3
Regards
Posted by swamib on Saturday, 07.22.06 @ 21:19pm | #58
Update for June - This sector based strategy had a small recovery in June, gaining 2.1% for the month while the market gained 0.2%. For the year this strategy has returned 7.4% while the market has returned 3.1 as measured by the SPY. The trading rules dicatate another trade as of market close today, selling Materials (XLB) and buying the Russell 2000 Value fund (IWN).
Posted by hmTodd on Monday, 07.3.06 @ 04:55am | #54
Update for May - This strategy led the market down in May, losing 5.0% while the broad market as measured by the SPY lost 3.0%. For the year the Sector Strategy has gained 5.2% while the SPY has gained only 2.8%. Going into June we will keep our position in Materials(XLB) but the model calls for closing out the position in Small-Cap Growth (IWO) and move into Energy(XLE).
Posted by hmTodd on Thursday, 06.1.06 @ 07:39am | #51
Update for April - This strategy had another positive return in April, besting the market 2.0% to 1.3%. For the year the Sector Strategy stands at 10.8% vs. 6.0% for the SPY. Going into May we will keep our position in Materials(XLB) but the model calls for closing out the position in Mid-Cap Growth (IWP) and move into Small-Cap Growth(IWO). IWP was picked up by the model in November and has returned over 14% since that time.
Posted by hmTodd on Sunday, 04.30.06 @ 13:56pm | #48
Update for March - This strategy did well in March, gaining 4.0% versus 1.7% for the market. For the first three months of the year this Sector Strategy has gained 8.6% versus 4.7% for the SPY. Going into April we maintain our positions in Materials(XLB) and Mid-Cap Growth(IWP).
Posted by hmTodd on Sunday, 04.2.06 @ 10:45am | #40
Update for February - This strategy suffered a bit in February with the sell off in energy, losing 5.7% on the month compared to +0.6% for the SPY. But for the year our strategy still leads 4.4% to 3%. We trade in our energy position (XLE) for basic materials (XLB) this month. I only hope it can be so successful. Although it cost us this month, our strategy has held XLE for almost two years and it has risen over 75% during that time. Not all bad!
Posted by hmTodd on Wednesday, 03.1.06 @ 08:33am | #34
Adam, you ask some good questions. First, the 2-week version backtests with a slightly lower CAGR than the 1-month version, but probably within the normal noise(randomness). It had 23 vs 17 trades for 2001 - current. I've found that the more often the re-assessment period the more stable the indicator needs to be, making a simple indicator like 6 month total return less useful. When you get down to daily or even weekly periods you need a more complex measure like regression.
I've tried numerous backtests. The ones I have published so far are relatively simple and limited to the stats available on the site pages. However, you might see others in the near future. I have not yet found a short screen that I like, and yes I've tried several.
Thanks for the comments and questions.
Posted by hmTodd on Thursday, 02.16.06 @ 18:46pm | #33
Couldn't help making a second comment! Have you tried shorting the two worst performing sectors each month (or every 2 weeks)?
Posted by Adam on Wednesday, 02.15.06 @ 19:02pm | #32
I love the simplicity of your strategy, but I'm curious what other strategies you've tested. Could you list them? You provide a large number of screening variables on your site and it's hard to believe that none of them beat the 6-month performance screen. Also, have you looked at the results and number of trades if you were to re-assess every 2 weeks rather than every 4?
Posted by Adam on Wednesday, 02.15.06 @ 18:58pm | #31
Update for January - Both of our published strategies began the year with exceptional returns. This Sector Strategy gained 10.7% in January compared to a 2.4% gain for the SPY. A solid gain for this strategy even if it did underperform the International Strategy. Again, no trades this month since both XLE and IWP are still in the top 6 positions of the screen ranking.
Posted by hmTodd on Wednesday, 02.1.06 @ 05:57am | #27
Paul,
Interesting that such a strategy would work with ETF's when it has supposedly not worked with mutual funds. But then he is eliminating the narrowest funds from his universe before he selects. That might be the key. What he is eliminating is exactly what our two strategies focus on. What does that say?
Hugh
Posted by hmTodd on Tuesday, 01.3.06 @ 13:18pm | #26
Hugh, thought you'd be interested in this
article. Particularly the system.
Paul
http://www.marketwatch.com/news/story.asp?guid={B369BBBB-7FAF-4E87-AD9A-46D91BD13A73}&siteid=yhoo
Posted by Paul Crum (rushes100) on Tuesday, 01.3.06 @ 09:49am | #25
Update for December - This Sector Strategy returned 1.4% in December compared to a loss of 0.2% for the SPY. For the year, the strategy is up 30.7% vs 4.8% for the broad market. Again, no trades this month since both XLE and IWP are still in the top 6 positions of the screen ranking.
Posted by hmTodd on Monday, 01.2.06 @ 10:41am | #21
jay,
Since I cannot give individual investment advice I will tell you my view on strategies like this, but you will have to determine your own actions.
These strategies provide me with general guidelines that I must put in context. Once in a strategy the rules are clear and my actions should be as well. Getting into a strategy I try to avoid putting myself at too much risk. Ideally I would jump into whatever the current holding are. But a few wks ago I purchased some international ETF's using a similar strategy but decided to avoid the top performer (Brazil, EWZ) because it had experienced significant price increases. That was a judgement call on my part that may or may not prove to be the best option. I used the strategy as a guideline but added judgement.
All that said, it looks to me that the top funds for this strategy today are the same as on my November update, XLE and IWP. I think the decision is pretty clear to me even though I've been worried about the energy stocks for a while and continue to think they are at risk. But so far I have been wrong.
Another option I've used when uncomfortable starting out is to begin with small positions and add money as time goes on. Hope this helps and good luck with your decision.
Posted by hmTodd on Wednesday, 12.21.05 @ 11:37am | #20
If and when I try to follow this system you are already in 2 positions, and then I get into it at the end of the month... what if 2 other etf's hold the top positions.
Would you say to get into the one's you are already in because they are in the top 6 or would you think it is okay to just buy the top 2 even if they are not the ones you are holding?
thanks
Posted by jay on Wednesday, 12.21.05 @ 09:11am | #19
My only knowledge of "Guy Lerner" system is in
the paragraph named "The Methodology". It looks
like he is using a determination of the "Bull"
and/or "Bear" market bias to "buy" new ETFs.
Other than that I don't know.
Paul
Posted by Paul Crum on Tuesday, 12.20.05 @ 08:32am | #18
Paul,
Thanks for the link. He used 131 days and I had used 126, not much difference. I want to do a little work with the etf's he picked and will get back with you. Any idea what his market timing strategy is?
Hugh
Posted by hmTodd on Monday, 12.19.05 @ 16:40pm | #17
In rereading your etf "investing concept",
which I think is great, I had come across this
post on the Yahoo board (it's actually from
the Street.com) that I thought Hugh would be
interested in.
Let me know what you think.
Paul (rushes100)
the link:
http://www.thestreet.com/pf/etf/guylerner/10249858.html
Posted by Paul Crum on Monday, 12.19.05 @ 11:34am | #16
pbd100 asked about using a 3-month lookback for this strategy instead of the 6-month. I just reviewed results from such a model and the 6-month lookback outperformed during the time period of this trial in total and for each year except 2004. In that year the 3-month lookback returned 20.4% while the 6-month returned 18.7%. For 2001 through Nov, 2005, the 3-mo had a CAGR of 1.8% compared to 14.6 for the standard model.
Another factor is number of trades. While the 6-month model had 17 trades through Nov of this year, the 3-month model recorded 42 trades. That is a surprisingly significant increase.
Posted by hmTodd on Monday, 12.5.05 @ 10:29am | #14
Have you looked at the same strategy with a 3 month look back instead of 6 month? I would expect a slight improvement in returns.
Posted by pbd100 on Friday, 12.2.05 @ 07:54am | #13
Update for November - This strategy returned 2.2% in November compared to 4.4 for the SPY. Year to date, the strategy is at 28.9% vs 5% for the broad market. No trades this month since both XLE and IWP are still in the top 6 positions of the screen ranking.
Posted by hmTodd on Friday, 12.2.05 @ 07:43am | #12
I won't argue with the logic of what you suggest. The problem is, as you mentioned, that only IYR has been around long enough for the backtest. Adding IYR in with the other funds does not improve the returns at all, in fact it is never selected over this short time period.
Posted by hmTodd on Wednesday, 11.23.05 @ 04:48am | #11
I like this approach but instead of all the Russell Funds I would look at ETFs that reflect major markets following the approach of John Murphy's Intermarket Analysis. What that means is I would have IYR (Real Estate), GLD (Gold), EFA (global markets), IWC (Russell Microcap) and some selection from IEF, TIP, & SHY (all Bond Funds). The problem with this is that some of these are new funds and would not fit into a backtest.
Posted by pbd100 on Thursday, 11.17.05 @ 22:50pm | #10
Update for October - As suspected this month turned out to be a poor one. Our two funds, XLE and XLU, lost 7.9% for the month handing us our 3rd worst performance since 2001. I'd feel bad about it, but as of the end of October these two funds were up 63% and 30%, respectively, since our model picked them. November began with one trade. Our utility holding (XLU) was sold and replaced with the iShares Russell MidCap Growth Index Fund (IWP). Looking for a better November,
Hugh
Posted by hmTodd on Friday, 11.4.05 @ 11:09am | #9
Update for September:
Since the model was holding energy and utility ETFs, I'm sure no one is surprised that the model had a great month in September with a 5.4% return, taking the ytd return to 36.9%. The SPY had a 0.8% return in Sept and 3.0% return for the year.
At this point it looks like October will be interesting. We've had some significant down months in the past, and this month is starting out that way. Still holding XLE and XLU through October.
Hugh
Posted by hmTodd on Thursday, 10.6.05 @ 10:49am | #8
Good screen that keeps you in the long term trend. Who would think it would be this simple? It just shows consistency pays off.
Posted by Trader on Thursday, 09.29.05 @ 08:10am | #7