We’ve made some changes to the new Chaikin Model in the P123 Models section.
We’ll detail the changes soon(added/changed some sell rules). In the meantime here are the improved results
Annualized return: 48% (up from 43%)
Sharpe Ratio: 1.52 (up from 1.32)
Alpha (since inception): 47% (up from 42%)
Alpha (last 3 years): 49% (up from 37%)
Realized Winners: 60% (about the same)
Here’s the link to the new model
There were two adjustments to the buy rules:
Use Rank>=90 instead of >=95.
An industry performance rule was also added that limits results to stock of companies whose industries ranked in the top 20% when it comes to 13 week average share price performance
One new sell rule was also added:
It forces the model to exit a stock that closes below the 90-day Chaikin Trend indicator and the 90-day Chaikin indicator is lower than where it was the prior day.
Why is not present any tradeability rules there?
Something like:
AvgDailyTot(50) > 500000 & Price > 2
It uses the Russel 2000 universe which should not have any liquidity issues. Feel free to run a sim using liquidity filters to see if it makes a difference and let us know.
If you start off with a few millions, some liquidity filters are probably necessary.
What am I missing? The stats page of this portfolio shows a YTD return of 5.16%, while the “performance by calendar year” show return in 2010 to be 0.61%.
Worse yet, the YTD performance in the weekly performance email shows 1 year performance to be 51.77%. When I rerun the portfolio for 1 year (as a simulation ending 7/30/10) the return is only 1.05%
Performance
Return Model S&P 500
Total 4,024.49 % -3.86 %
Annualized 48.98 % -0.42 %
Year To Date 5.16 % -1.21 %
4 Week 19.37 % 7.73 %
13 Week -3.52 % -7.17 %
1 Year 51.77 % 11.64 %
3 Year 213.04 % -24.30 %
Annualized Performance by Calendar Year
2001* 2002 2003 2004 2005 2006 2007 2008 2009 2010**
Model 3.99 % 9.25 % 150.03 % 56.60 % 37.88 % 72.70 % 46.35 % 26.20 % 102.37 % 0.61 %
S&P 500 -1.38 % -23.37 % 26.38 % 8.99 % 3.00 % 13.62 % 3.53 % -38.49 % 23.45 % -2.14 %
Excess Return 5.38 % 32.62 % 123.65 % 47.60 % 34.88 % 59.08 % 42.82 % 64.69 % 78.92 % 2.75 %
33% of annual data is required to calculate annualized return.
(*) Inception Date: 03/31/01
(**) End Date: 07/30/10
aposner,
The 5.16% YTD is correct. The 0.61% for 2010 in the Performance->Stats I’d have to investigate. It is a different logic (annualized), but it does seem to be off.
Running a 1Y sim is not the same. 1Y ago the port had running positions that were obviously different than starting a sim 1Y ago: they were bought in previous rebalances, and held on. It is a big difference, but with only 10 positions a couple of differences can compound dramatically.
The 0.61% for 2010 in the Performance->Stats was ignoring the last data point. Should be corrected soon.
Does this model still exist? I see a chaikin ranking system but nothing else.
The link does not work for me
It’s no longer a Portfolio123 strategy, but there are a number of public versions you can play with. Click on Research, then Research search, then put “Chaikin” in the “model name” field and you’ll see them.
The ranking system was created by Marc Chaikin himself. See https://www.portfolio123.com/rank_about.jsp?rankid=161238
Do we know why this particular one was removed?