Strategy Development
Stockscores.com Perspectives for the week ending July 8, 2007
In this week's issue:
As a trader and market analyst, I try to find stock market patterns that are predictive. When I first started researching the stock market nearly 20 years ago, I would look at the chart of every stock that had performed well. I wanted to see if there were common traits at the start of their up trends that I could watch for when trying to find good stocks.
This approach makes good logical sense but there are problems that need to be overcome. First, we need to know that the common patterns that seem to occur at the start of every up trend are in fact unique to stocks that are likely to go up. It would be silly to say that all stocks that I found to go up had traded between $1 and $20 and therefore buying any stock between $1 and $20 will result in success. While it may be true that all the stocks that I considered were between $1 and $20, it can hardly be said that it was this criteria that led to the upward trend.
So, we need to take out the randomness of events and ensure that there is a statistical correlation between the pattern we see and the result. To do that, we must test not only the stocks that have done well for the pattern but also stocks that have done poorly to ensure that the pattern does not also repeat itself with a negative result.
That requires taking a random sampling of stocks and testing all for the pattern to see what the result is. For example, I have found that abnormal price and volume activity combined with the elements of good chart patterns are useful for predicting upward trends. To study this, I tested the criteria against a random sample of 50 stocks. I printed out the charts showing some proprietary indicators that I use to test for abnormal activity and then tested to see what the resulting trend was after my criteria were met. It was from this process that I found a correlation between my criteria and performance that was statistically significant and not just random.
However, this process did bring up another problem that had to be overcome. Historical patterns look very different after an up trend than on the day when the trade would be considered. When a stock goes in to a strong up trend, the scale of the chart changes and everything gets squished lower. A "good" chart pattern tends to look that way with the benefit of hindsight. Sample bias is another factor that needs to be overcome.
So, chart patterns had to be observed based on the way they looked on the day of the abnormal activity. I had to print out the charts as of the abnormal event date and compare the patterns with stocks that had the same abnormal event but without the resulting success. This process ensured that my judgment of the chart pattern was based on what I saw and not the ultimate performance.
The final step is to determine the expected profitability of a strategy based on the pattern that you can test. It is not about being right or wrong, it is about how much you make when you are right versus how much you lose when you are wrong. A strategy is only effective if the expected value of the trade is positive, meaning the profits outweigh the losses over a large number of trades.
The summer tends to be a slower time in the market and a good time to do this sort of research. Observe stocks that have done well and try to figure out if there are recurring patterns that predict their strong performance. Write down the strategy rules and test them against a random group of stocks. Ensure that you don't have a research bias by omitting the performance result from the stock chart you are measuring. Ultimately, you need to ensure that there is a strong correlation between the pattern you can identify and money making results.
Back To Top
The market was very slow this week because of the holidays in both the US and Canada. That meant low trading volumes and not a lot of opportunity. For our daily newsletter readers, I selected two stocks on Thursday that I show below:Back To Top
1. CKSW
CKSW achieved a new closing high after about five months of sideways trading. Price and volume action was abnormal today as the stock breaks out from low volatility, sideways trading. Support at $3.60. This stock made pretty good gains again on Friday so it may be better to consider this stock on a pull back.
Back To Top
2. RYN
RYN breaks through very long standing resistance out of an ascending triangle pattern that has been building over the last year and a half. Support at $44.80.
Back To Top
References
Get the Stockscore on any of over 20,000 North American stocks.
Background on the theories used by Stockscores.
Strategies that can help you find new opportunities.
Scan the market using extensive filter criteria.
Build a portfolio of stocks and view a slide show of their charts.
See which sectors are leading the market, and their components.
Disclaimer
This is not an investment advisory, and should not be used to make
investment decisions. Information in Stockscores Perspectives is often
opinionated and should be considered for information purposes only. No
stock exchange anywhere has approved or disapproved of the information
contained herein. There is no express or implied solicitation to buy or
sell securities. The writers and editors of Perspectives may have positions
in the stocks discussed above and may trade in the stocks mentioned. Don't
consider buying or selling any stock without conducting your own due diligence.
Back To Top
|
If you wish to unsubscribe from the Stockscores Perspectives Weekend Edition or change the format of email you are receiving please login to your Stockscores account. Copyright 2006 Market Perspectives Inc.
|
|
