No one mentioned www.canslim.net I have never used it but some have mentioned it in the past in regards to the canslim methodology.
Right. It is not strict CANSLIM. But one of the critical factors of CANSLIM is the patterns (such as cup with/without a handle). If we can test this pattern, and know the performance statistically without even considering the fundamentals, I think it is valuable. If the test results shows CANSLIM doesn't work, it doesn't mean CANSLIM not work because lack of fundalmentals in the test. If the test results show the patterns used by CANSLIM work even without qualitative/fundalmental elements, I believe CANSLIM should do even better. The test only tests the past not the future. It only proves by quantitative analysis based on historic data. It is up to people to interpret the results.
However, if you're going to test the patterns alone, you're going to have to be very specific about the characteristics of those patterns. How will you avoid encountering the problems that others have run into trying to do the same thing, e.g., Bulkowski? For example, how much latitude will you allow in the depth of the cup? Will you require that the cup occur in an uptrend? If so, how will you define "uptrend"? Will you require a volume dryup in the bottom of the cup? If so, how dry? And so on.
Yes. The definition for patterns used in the study is parameterized with specific values such as the width and the depth of the cup on a uptrend. With the help of AQ software, it is easy to complete and doing such things. For example, to find the tops and bottoms of points skipping small waves. I also visually checked many of the patterns found. The recognization is good. Of course there are some rare cases which are not very good. If there are enough patterns found as did in the study, the test result should be used in the quantitative analysis. I didn't consider volume dryup in the bottom of cup and it should not affect the test much. Maybe I can parameterize volume dryup sometime later by defining the average volume around the bottom decreasing as compared with the average volume of the left side of the cup. There are places to improve the study. Thanks.
The crux of CANSLIM is not cup and handle. It is basically a earning and price momentum based strategy. Also cup and handle is not the only pattern WJO talks about. He mentions several patterns Cup with handle Saucer with handle Double Bottom Flat base High, tight flags Base on top of base Ascending bases. CANSLIM is a very popular methodology amongst many hedge funds and institutions. For them WJO has a different product line called WONDA. With WONDA and other services of WJO what you get is lot different from IBD. There are couple of other methodologies very similar to CANSLIM or improvement on CANSLIM which are also popular with institutional investors and hedge funds. I think your study has fundamental flaw in understanding the methodology itself. http://www.winninginvesting.com/canslim.htm CANSLIM = Current Quarterly Earnings Look for companies with the largest increases in quarterly earnings compared to the same quarter last year. Bigger is better! Seventy-five percent of the winners in OâNeilâs study had quarterly earnings jumps of 70% or more. The minimum acceptable increase is 18%. One caveat though, avoid stocks with tiny year-ago earnings, e.g. less than five cents. Huge increases donât mean much if you're comparing to a miniscule year ago number. CANSLIM = Annual Earnings Consistent annual growth rates of 25% or more over the last four or five years are an important factor. Consistency is the key word. Make sure forecast earnings for next year are in line with the historical growth rate. CANSLIM = New Products or New Management, and new Highs Look for a catalyst that will propel fast earnings growth. O'Neil also includes new highs in stock price here. He wants stocks at or near their all time highs. This is the hardest aspect of the strategy for most new investors to accept. Our natural tendency is to buy stocks after they have gone down in priceânot when they are at new highs. CANSLIM = Supply of Stock Stock prices move as a result of supply and demand for the companyâs shares. If thereâs not many shares in circulation, a small amount of buying could push prices up quickly. Look for companies with 5 to 25 million shares outstanding. CANSLIM = Leader Here OâNeil is talking about stock price action, not a companyâs success in selling product. Look for stocks that have outperformed at least 80 percent of the rest of the market during the past year. The company should have the best performing stock in its industry. CANSLIM = Institutional Ownership Institutions are mutual funds, corporate pension plans, insurance companies, etc. OâNeil likes some, but not too much institutional ownership. Look for 5% to 25% institutional ownership. CANSLIM= Market Direction Very few stocks go up when the market is going down. Buy only when the market as a whole (S&P 500 Index) is going up. Those are the selection criteria. Once youâve picked promising candidates, you have to decide when, if ever, to buy the stock. OâNeil looks at a stockâs price chart for guidance. He looks for stocks that have been consolidating (bouncing around a limited range) for a while, and then move up to (or close to) new highs.
Unfortunately, not one study or portfolio have ever followed the criteria. Which pretty much takes us back to where we began . . . :eek:
Maybe the study should be renamed to be "study of some patterns from CANSLIM". The test doesn't mean to prove wrong. If the pattern itself works statistically, CANSLIM should do much better after considerations of all other factors. Why is there a pattern? It must has its reasons as described in the book by O'Neil. I believe patterns actually already embbeding some fundalments in it to some extend in most cases. Any test has its limitations as pointed out in my study. If you claim XYZ work, someone proves X without YZ also works. It means XYZ together should do better. Thanks.