A Simple, Efficient and Convenient Universal System

Vincent Granville, Ph.D., 
Data Shaping

We propose an original system that provides reliable daily index and stock trending signals. The non-parametric statistical techniques
described in this article has several advantages: simplicity, efficiency, convenience and universality.

SIMPLICITY:
There are no advanced mathematics involved, only basic algebra. The algorithms do not require sophisticated programming techniques. They rely on data that is easy to obtain.

EFFICIENCY:
Daily predictions were correct 60% of the time in our tests. This good performance can be improved using techniques described in this article.

CONVENIENCE:
The system does not require backtesting. It is parameter-free, and does not need periodic updates. Additionally, the algorithms are very light in terms of computation, providing forecasts in a snap even on very slow machines.

UNIVERSALITY:
The system works with any stock or index with a large enough volume, at any given time, in the absence of major events impacting the price. The same algorithm applies to all stocks and indices.

ALGORITHM
The algorithm computes the probability, for a particular stock or index, that tomorrow's close will be higher than tomorrow's open by at least a specified percentage. The algorithm can easily be adapted to compare today's close with tomorrow's close instead. The estimated probabilities are based on at most the last 100 days of historical data for the stock (or index) in question.

The first step consists of selecting a few price cross-ratios that have an average value of 1. The variables in the ratios can be selected so as to optimize the forecasts. In one of our applications, we have chosen the following three cross-ratios:

Ratio A = ( today's high / today's low ) /
( yesterday's high / yesterday's low )
Ratio B = ( today's close / today's open ) /
( yesterday's close / yesterday's open )
Ratio C = ( today's volume / yesterday's volume )

Then each day in the historical data set is assigned to one of 8 possible price configurations. The configurations are defined as follows:

Configuration 1: Ratio A > 1, Ratio B > 1, Ratio C > 1
Configuration 2: Ratio A > 1, Ratio B > 1, Ratio C <= 1
Configuration 3: Ratio A > 1, Ratio B <= 1, Ratio C > 1
Configuration 4: Ratio A > 1, Ratio B <= 1, Ratio C <= 1
Configuration 5: Ratio A <= 1, Ratio B > 1, Ratio C > 1
Configuration 6: Ratio A <= 1, Ratio B > 1, Ratio C <= 1
Configuration 7: Ratio A <= 1, Ratio B <= 1, Ratio C > 1
Configuration 8: Ratio A <= 1, Ratio B <= 1, Ratio C <= 1

Now, to compute the probability that close tomorrow will be at least 1.25% higher than tomorrow open, we first compute today's price configuration. Then we check all past days in our historical dataset that have that configuration. We count these days. Let N be the number of such days. Then, let M be the number of such days further satisfying the following:

Next day close is at least 1.25% higher than next day open. The probability that we want to compute is simply M/N. This is the
probability, based on past data, that close tomorrow will be at least 1.25% higher than tomorrow's open. Of course, the 1.25 figure can be substituted by any arbitrary percentage.

PERFORMANCE
There are different ways of assessing the performance of our stock trend predictor. We have investigated two approaches:

1. computing the proportion of successful daily predictions, using a threshold of 0% instead of 1.25%, over a period of at least 200
trading days
2. using the predicted trends (with threshold set to 0% as above) in a strategy: buy at open, sell at close or the other way around based on the prediction

Our tests showed a success rate between 54% and 65% in predicting the Nasdaq trend. The strategy associated with the forecaster has been analysed on our web site. Check our section on universal keys at http://www.datashaping.com/buy.shtml

Even with a 56% success rate in predicting the trend, the long-term (non compound) yearly return before costs is above 40% in many instances. Note that we provide similar strategies that do not rely on the open price to interested clients. As with many trading strategies, the system sometimes exhibits oscillations in performance. It is possible to substantially attenuate these oscillations, using a technique described on our websiet: http://www.datashaping.com/newsletter083101.shtml

In its simplest form, the technique consists of using the same system tomorrow if it worked today. If the system fails to correctly predict today's trend, then use the reverse system for tomorrow.

UNIVERSAL FORECASTER
Universal Trend Forecaster is the full name of our implementation of this system. It is available online, at http://www.datashaping.com/members.shtml

You can check out the real past performance (last 365 days) online, for any stock or index, by entering the stock symbol in the trading box and clicking on the submit button. Additionally, we plan on releasing an Excel template containing all the formulas to perform the required computations.