**as of 11/18/2009**
nice! where do i sign up?
Interesting index, wondering what was used to hedge the index?
hi alex thanks! i will be offering it for free for some time after that it will be on subscription.
James, thanks, i just used the QQQQ to hedge the index (100% short the QQQQ at all times and 100% long the index) obviously hedging using the bottom rated stocks did much better with less risk.
By performance, does it mean holding long these stocks for the entire duration and shorting the QQQQ ? or is it applying some other (for example short term mean reversion) strategy to the longs..
in that case, I wonder if there is an impact due to survivorship bias especially in the ndx100 since many of ndx100 of 2000 do not exist now.
Another issue is that stocks like GOOG for instance did not exist since 2000 so that will skew the results as well..
sp500 may be a good candidate to test on ..
hi badri, actually in this case we rebalance weekly, and the turnover is a little lower than standard momentum strategies (average hold of about 3 months). As for survivorship bias, over this particular time frame it had a slight positive impact, and in testing on the S&P500 we got even better results than this.
How often does the composition of say the “Top 10” change? Is based upon a periodic review?
hi, it is rebalanced weekly. cheers
How did you deal with survivorship bias?
I am guessing that the selection function considered price as one input to sort the stock list.. in which case, automatically the losers fall off during rebalancing.. (they are no longer in the top N) but this is a guess.
hi damian—this is why i built my own database in this case versus research wizard. We also manually input an actual list of the 100 stocks in the index at the time each quarter classifying each as fitting the criteria for the index or not to compare results with using the current 100.
I knew you would ask! :o)
David, Re: reply to Badri. (“testing on the SP500, we got better results”)Does that mean using the 10 from the Nasdaq and hedged with the SPX , or using instead 10 from the SP500 in lieu of your Nasdaq Livermore?
hi, yes indeed we got higher results (close to 100%) on the S&P500.
If the SP500 version of the Livermore is much better (close to 100% better), I don’t really understand
why do you then use the Nasdaq version of the Livermore.
Please, use and publish the SP500 version.
That’s probably the most alpha I’ve ever seen. For comparison of similar “stock picking screening” approach, I ran the numbers on the 61 screens over at AAII, using the same time period. The best performing strategy is CANSLIM with a 17.7% CAGR. This Livermore index absolute destroys any long/short systems out there.
Is there any assumption about transaction costs here? What’s the assumption about when the index is bought/sold? What day of the week and open/close?
can you please provide some details on the weighting of the portfolios? For example, is the 10-stock portfolio equally weighted, market cap weighted, vol weighted, etc. In addition, is the hedge position in the QQQQ equal in dollar terms to the 10-stock portfolio?
hi JJ, it is actually an equal weight applied to each…..obviously dynamic weightings help a lot.
How much of the performance is from the Top 10 and how much from the hedge?
The hedged strategy is especially attractive with the low DD. Have you given a thought to dynamic hedging, i.e. you would be 100% hedged when the market is getting overbought based on some of your other indicators, and lower hedging for oversold markets? Would be neat to see the results of such a combination.
hi andras very good question…..yes of course, and we will be showing the impact of dynamic hedging on the portfolio soon. It makes little sense to short the nasdaq in bull markets as a hedge.
Nevermind, I see it now at the top of the graphs!
Doesn’t this break Durable Concept #3, only use relative strength above 200-day MA?
hi, that concept still applies, however, this is not a pure high relative strength model…….there is more to it than that! that is why it performs well it more than just the 200 ma environment.
I was wondering (your time permitting) if you could show results for Top5 or Top1 hedged…or even rebalanced biweekly or monthly for those of us with busy lives? : )
David (and all),
Intrigued by your index and above all by the astonishing performance, I attempt to understand what could hide behind that Livermore Index. To guess your filter and try to replicate the findings of your Top10, I tried to read between the lines. Here my guesses and approach (which was initiated by luck, but as said Edison inventions are 70% made of luck):
1) What is a strong stock, I guess one that does not correct much while always going higher: the recent performance of a stock for a break-out system (à la O’Neill, Livermore or Darvas 😉 with a close stop and an exit with a low DV2 should tell you about that => My first indicator
2) You mentionned volumes… A strong stock always climbs on more volume than during its corrections. What about a ratio of the sum of the volume of positive days by the sum of the volume of the negative days => My second indicator
3) I want a minimum of volume, i.e. minimum daily out of x days > 300,000
4) Like O’Neill and Livermore, let’s not trade stocks in the low figures. Minimum = 10$
5) I combine the 2 indicators, include filters for volume and price and take the bests of the class.
Here my picks:
1) Nasdaq: HGSI, BEXP, SVNT, VMED & ISRG
2) NYSE: UFS, OSK, GOL, SOA & VCI
Hi David: will you be publishing updates for the Index – you mentioned weekly rebalancing and the last date is 11/18?
hi adam, we will be previewing a daily publication on the livermore index shortly—it will be offered for a month and will contain other analytics.
Fill in your details below or click an icon to log in:
You are commenting using your WordPress.com account. ( Log Out / Change )
You are commenting using your Twitter account. ( Log Out / Change )
You are commenting using your Facebook account. ( Log Out / Change )
You are commenting using your Google+ account. ( Log Out / Change )
Connecting to %s
Notify me of new comments via email.
Enter your email address to follow this blog and receive notifications of new posts by email.
Join 1,097 other followers
Create a free website or blog at WordPress.com.