Hey Fools,I have spent a lot of time recently contemplating the recent volatility in tech stocks, and examining the methodology used to develop the highly successful screens discussed on the Foolish Workshop and Mechanical Investing Boards:http://boards.fool.com/Message.asp?id=1030063000770000Most of the screens created there have very, very, good track records over the course of this bull market, even though they are all pretty much suffering at the moment with the recent pullback in the Nasdaq (with the notable exception of Peg26!):http://www.fool.com/Workshop/WorkshopReturns.htmHowever, my KEY-TSE is suffering the downside even more seriously, after a brilliant beginning in it's first quarter of existance. Clearly, this is a very volatile way to invest. Only the future will determine the value (or lack thereof) of this creation of mine.I have come up with a couple new ideas for screens, after examining all of the screens covered by Jack Cade in his excellent weekly post of rankings, which includes a brief description of the method used for each screen:http://boards.fool.com/Message.asp?id=1030013012206000&sort=idClearly the best screens benefit from the use of The Value Line Investment Survey, which unfortunately doesn't help us develop screens in Canada. The foundation for MI screens is Value Lines timeliness rankings, which rank stocks from 1 to 5, based mostly (but not totally) on recent stock market performance. Value Line timeliness=1 ranking includes 100 stocks, and t=2 includes an additional 300 stocks. I have tried to get some improvement in screen performance with a screen i'm calling KEY-ANNUAL by starting with the 200 best performing stocks over the last year, and with a new monthly screen i'm calling RS13-M, by starting with the 100 best performing stocks over the last 13 weeks. Here is what i've done:KEY-ANNUALAll screens make use of the Globe and Mail's filtering functions at their Globeinvestor site:http://www.globeinvestor.com/static/hubs/filters-historical.htmlUsing the historical filter, select Industry=all, Security=common, Exchange=Toronto, Price greater than $5, 1 Year change greater than 18%.Adjust the 18% return setting until globeinvestor selects as close as possible to 200 stocks. This naturally will depend on current market conditions.When you have a return setting that will identify just over 200 stocks, click on the "Financial Reports" link at the top of the list. Next, click on the "Market Cap" column heading, which will sort the list by market capitalization.Print out, or write down, a list of the top 30 stocks by market cap from this list. Next is the only step that requires a bit of time and effort. According to the MI Keystone screen description, the top 30 by market cap need to be sorted by 26 week return. This time period shows up again and again as a highly successful length of time for ranking stocks for relative strength. Unfortunately, Globeinvestor doesn't offer this time period for sorting stocks. Therefore, I've taken matters into my own hands, so to speak.With your list of 30 good stocks, go to Yahoo's Historical Quotes page and enter the ticker symbol, with .to added at the end. Nortel would be nt.to and Bombardier would be bbdb.to (for ticker bbd.b). Enter in the current date and the date six months ago, which will give you all closing prices for that time period.Yahoo's Historical Quote service:http://chart.yahoo.com/dA bit of math needed here: (Today's Price-6 Month Old Price) divided by 6 Month Old Price. Multiply by 100 and you have 26 week return. Note: Use far right hand column "Adj. close", since this adjusts for splits and dividends.Find the top 5 stocks, out of your list of 30, with the highest 26 week return. Buy equal dollar amounts of each of these 5 and hold for one year. After a year, sell these and buy the newest list of 5 in equal dollar amounts by running this procedure again.I will be tracking the picks chosen by this screen on April 28 until the end of the year. At that time I will run the screen over and rebalance at the start of 2001. This is because bdfinney's MI backtester shows that the Keystone screen performs the best when rebalanced and started early in January. A "January Effect" at work here."The Backtest Tool" at bdfinney's site:http://www.phillysites.com/screens/backtest.asp?s=keyshows that the MI Keystone screen, 5 stock, annual hold, has made a compound annual return of 27.8% from January 86 to January 98. At that rate of return, $10,000 invested January 86 would have been worth $310,000 in January of this year, just 14 years later.RS13-MNow that that long winded description is out of the way, here is a simpler screen, based of course on the RS13 MI screen:Using globeinvestor, historical filter, select Industry=all, Security=common, Exchange=Toronto, Price greater than $5, 1 quarter change greater than 20%.Adjust the 20% return setting until globeinvestor selects as close as possible to 100 stocks. This naturally will depend on current market conditions.When you have a return setting that will identify just over 100 stocks, click on the "Financial Reports" link at the top of the list. Next, click on the "Market Cap" column heading, which will sort the list by market capitalization.Select the top 3 (or 5 if you wish) stocks from this market cap sort, and purchase in equal dollar amounts. Hold for one month and re-run the screen. Hold all shares of any stock still on the list, selling all others. Split the proceeds evenly and purchase the new stocks on the list.One problem that I found with the KEY-TSE screen procedure was that I was always starting with a different sized group of stocks. With that method calling for settings of 1 year>50% and 1 quarter>20%, you would be starting with a smaller and smaller list of stocks as the market returns slow down. I believe this new method of finding a starting list that is always the same approximate size, will be more consistent and reliable.Only time will tell, since we don't have the benefit of backtesting. I will be comparing these new screens with the KEY-TSE, TSE300, S&P500, and Nasdaq, going forward.I am no longer planning to track the RS-TSE as I believe it would be way too volatile for anyone to actually stick with, long term.Any and all comments, criticisms, and suggestions are gladly appreciated. Here's to hoping for great returns for all in coming months!Crapshooter
Best Of |
Favorites & Replies |
Start a New Board |
My Fool |
BATS data provided in real-time. NYSE, NASDAQ and NYSEMKT data delayed 15 minutes.
Real-Time prices provided by BATS. Market data provided by Interactive Data.
Company fundamental data provided by Morningstar. Earnings Estimates, Analyst Ra