Instablogs are Seeking Alpha's free blogging platform customized for finance, with instant set up and exposure to millions of readers interested in the financial markets. Publish your own instablog in minutes.
-
Instablogged Stocks
Stocks that instabloggers have most recently written about -
Latest Instablog Posts
-
1
UPDATE: Afferro Mining Results Underline Its...
-
2
Falcon Oil & Gas Begins Drilling In Hungary
-
3
Shanghai Gold Premium Hits New High
-
4
Today's Technical Outlook- 5/23/2013
-
5
3 Reasons To Avoid Panic Selling
-
1
-
Top Instablogs
See all Top Instablogs »
A low drawdown strategy for sector rotation for Fidelity Select Funds 40 comments
The following strategy works quite well in backtesting, somewhat better than many other methods.
Backtesting yields the following results for the period 1991-2011 (All results are based upon the weekly adjusted close data from finance.yahoo.com. I will be glad to collaborate if a reader wants to repeat the calculations with data that he thinks to be more reliable.):
- CAGR 25%
- Number of years of losses: 2 (out of 21, in 1994 and 1998)
- Maximum loss: -11% (in 1994).
- Standard Deviation of Annual Returns: 22%
- Mean of Annual Returns: 27%
- 1, 3, 5, and 10 Year CAGRs: 4%, 27%, 24%, 28%
- 2011 Return: 4%
- Number of years the strategy beats SP500 index: 17 (out of 21).
- Mean of positive returns: 28%
- Mean of negative returns: -1%
- 95% Confidence Interval for Alpha (%):[8.7,33] (between 8.7 and 33)
- 95% Confidence Interval for Beta :[-0.2,1.1]
- 90% Confidence Interval for Alpha (%):[10,31]
- 90% Confidence Interval for Beta :[-.1,1.05]
(For the last four computations, annual ten year treasury rates were used as the risk free rates, and the annual returns of the SP500 index were used as the market returns.)As far as can be ascertained, by most measures this strategy outperforms all the publicly available strategies for investing in Fidelity Select Funds.
This strategy starts with FBIOX as the position in the first 13 weeks of the year 2012. We will update it every 13 weeks.
For comparison purposes, here are some statistics for the whole basket of Fidelity Select funds (each statistic is computed for the life of each fund during 1991-2011):
This is not investment advice, but just an effort to track the results of hypothetical investments according to a particular strategy.
I am not associated with Fidelity funds in any capacity except as an account holder.
Instablogs are blogs which are instantly set up and networked within the Seeking Alpha community. Instablog posts are not selected, edited or screened by Seeking Alpha editors, in contrast to contributors' articles.
Share this Instablog
This post has 40 comments:
I followed one touted newsletter using Fidelity. It "switched" every six months. In 2008, my indicators went to cash and he got clobbered by NOT doing so ... http://bit.ly/wQfHcd
2. This method made 12% in 2008 since it went to TLT in fourth quarter of 2008. Most other Fidelity trend following methods seem to have lost a lot in 2008.
It might be difficult for me to replicate the switching from VUSTX to TLT, but I'm going to give it a shot.
Good stuff Varan. I'm looking forward to working with your strategy.
http://bit.ly/whnqlP
Most of the other methods did not have a good return during the last ten years. This one yields uniformly good annualized returns: for any three year period the annualized return is in double digits, and except for two years, always greater than 15%. Very few trading strategies can achieve that.
the "top" 2 funds per period ( for a little diversification ). And for the present period what would be the second one behind FBIOX ?
The second fund for the beginning of the year comes out to be FSHOX. The re-balancing occurs on 4/6/2012.
I found your strategy heavily relies on
"Rank the performance of the funds during the period of 8 weeks ending on the close of the previous week."
I have tried to use last Q performance for picking the next Q winner, and the result is much worse.
It makes me wonder, why 8 weeks? how do you get this magic number?
2. Now make sure that if you change these by one or two weeks the performance does not change drastically.
and what are the CAGR?
ETF Paired Switching:
Ranking period: First full week of quarter to end of quarter.
Investment period: First DAY of subsequent quarter to end of quarter.
======================...
Fidelity Selects:
Ranking period: Last 8 weeks of calendar quarter
Investment period: One week exactly after ranking period ends thru 13 weeks..
Thanks
By the way, YTD returns have not been too shabby, either for the single or the two fund portfolio.
One Fund 12.5% YTD
compare with:
http://bit.ly/MT1Nms 6.5%
http://bit.ly/GYKJGV 8.9%
Spreadsheet: http://tinyurl.com/6oj...
yes, someone even convinced to me to lose some money on TBT to bet against TLT.
Nice to see someone following this. Are you actually using this strategy? Just curious. I have been using it since January.
3 and 5 year rolling returns have averaged 90%+ for the 3 year, and
200% + for the 5 year (back to 1991 ) .. amazing for something so simple and with a robust backtest sample size ! Still would like to diversify with at least 2 funds though ....
I have also cooked up a strategy based on risk parity and relative strength for the reasons that you mention (you can put only so much of your portfolio in one or two funds). It yields around 15.5% annually for the period starting in 1991 and entails allocating the portfolio (generally unequally) among fifteen top funds, and updates every two months.
In all these strategies, the key of course is to add some assets to the select funds whose returns are negatively correlated with the returns of the select funds. That part of the strategy is missing from most other systems as far as I can tell.
I use a similar strategy that I learned from an investment class. Using Fidelity Select Sector funds ranked on an annual basis choose the top fund to invest in for the next year. However, there are some simple tests to pass to keep this top ranked fund. If it was in the top 10% two years in a row pass over it and go to the next one. Also, if funds from the same sector are in the top 10% 2 years in a row go to the next fund on the ranked list. Have you thought of incorporating a similar filter for your method?
No, I have not attempted to incorporate such a filtering mechanism. It may be worthwhile to look into it.
http://tinyurl.com/9l6...
Since 4 of their funds dont go back far enough, I set this up and tested from 2001 on. Used VUSTX the whole way through. I get a big loss in 2008 which you didnt state in your post. I do get 24.73% CAGR but the system isnt very stable. When I choose a rebalance period of 63 days I get 25%, 64 days is 6.6% for example. Also, I am not synchronized to the calendar year as you are but in my opinion that should not be a requirement for a stable system. What do you think?
Sorry your post slipped my mind. I agree that it should not make a difference, but some time this kind of instability that occurs due to some discrete event (distribution for example), cannot be avoided. I have done some robustness calculations with weekly data but not with the daily data.
YTD return 20%
FBIOX 1/6 thru 4/6 16.47%
FSRPX 4/6 thru 7/6 -3.33%
FBIOX 7/6- to date 6.56%
Bench Mark Equally Weighted QQQ, SPY and DIA: 17.4% YTD
On 10/5/12 the system switches to FSAGX.
I will follow the strategy to the letter and switch to FSAGX at close tomorrow.
Thanks
Returns through first three quarters of 2012: 20.76%
Return during the last quarter of 2012: -15.74%
Total 2012 Return: 1.74%
YTD 2013 Return 12%
There are two ways to deal with the dismal returns of 2012:
1. Accept it, since despite the poor performance for 2012, the CAGR has been more than pretty good: 1991-2013 23%
2000-2013 24% , 2008-2013 19% (annual return for 2008 was 12.8% - postive).
2. Modify the strategy by excluding FSAGX from the basket. That improves the performance substantially, even for 2012.
Thanks for your interest.
Latest Followers
Latest Comments
Most Commented
Posts by Themes