Momentum/Trend Following with ETFs?

Discussion in 'ETFs' started by billpritjr, May 7, 2006.

  1. Wanted some opinions on the below strategy. This can be done with any account balance, but lets say $100,000 cash balance to start.

    Trader uses below data, and using "best performer for one month" information, keeps his 100K invested in the top 2 performing ETF's at all times. Trader could also use 90-day data.

    Trader could choose to limit his universe to whatever trading volume ETF, say those which trade 1M or more a day, for liquidity.

    SEE:

    http://screen.morningstar.com/etf/Lists/ETFReturns.html

    (sort 1-month return column)

    also:

    http://stockcharts.com/webcgi/perf.html?GLD,EWZ,EWT,OIH,EEM,RKH,EWS,EWY,EWW

    (adjust scroll bar to whatever time period)

    As we know, trend following does not predict, it just responds, but we also know the #1 ETF will not drop to #50 in one month. Things happen gradually. We also know that things in motion tend to stay in motion.

    Above strategy would require the trader to buy/sell ETFs every month to few months to remain in the top performer.

    Thoughts?

    Not saying this is the Holy Grail or even a good idea, just wanted to get some discussion going.
     
  2. you're going to be screwed if they both enter directinless phases at the same time :)

    ....and only holding 2 commodites, i don't really liek your chances very much.
     
  3. agpilot

    agpilot

    Hello Bill
    Back about 5 to 10 years ago on the FastTrack forum there were a few that did what you are suggesting. I think I remember that in back testing it was thought that having 3 positions gave the highest return but also the biggest drawdown. Having at least 5 positions gave a good return with quite a bit less drawdown. Having at least 10 positions gave even less drawdown but fair amount less on performance. Another important part was what they used to trigger a sell or switch and I regret I don't remember that but I believe it was a percent drop rather then a fixed time like your 30 day comment. It can work but you must remember that this was in the 1995 to 2000 timeframe and everything was going up forever and ever.
    Hope this helps. Run a search for FastTrack data service and email them for the latest location of the forum.
    agpilot
     
  4. I didn't know fasttrack ever had a forum?
     
  5. Several years ago TASC ran an article about a system that followed a similar strategy using Fidelity Select sector funds. As I recall they took the four with the biggest momentum from the previous month. Results were decent, but I don't remember the details.
     
  6. "It" has potential but it's more important to control your risk just in case the number 1 & 2 performing funds you're in become number 49 & 50 during a big intra-month price swing.
     
  7. agpilot

    agpilot

    To 99atlantic:

    FastTrack... full name: Investors FastTrack
    Website: http://www.fasttrack.net/ 4 the home page.. Note their data is for ETF, MFds, or stocks.

    Like any other method, the person using it can make or break the normal results others get...... agpilot
     
  8. ? about something similar - would it make sense to buy the top 2-3 largest discounts per sector and then drop them when they change +/- 1 % (in terms of the discount percentage)?

    I know it's different because most people would buy top 2-3/sector based on nav/market return, but what about based on discounts. Hmmmmmm.
     
  9. Discounts to what? What securities are you talking about?

     
  10. I think he must be talking about closed end funds, not ETF's.
     
    #10     May 11, 2006