StockEval News

Please download and run QuanTek on your Windows computer, by going to the Download QuanTek page. A description of the most important features of QuanTek is given below, as well as new features added in subsequent updates and a summary of future plans.

Future Plans

Our next project for the QuanTek program is to implement the Capital Asset Pricing Model (CAPM), and at the same time a multivariate GARCH model for the Price Projection. GARCH stands for Generalized Auto-Regressive Conditional Heteroscedasticity. In other words, the time-dependent variance is taken into account in the Price Projection. This requires a multivariate approach to predicting the future returns, rather than the univariate approach used now (only past price data are used in the prediction of future price data). The Optimal Portfolio calculation will similarly be updated to incorporate this new Price Projection as well as the CAPM.

For the present, the Wavelet filters assume a stationary process, by averaging over all the time values of the Wavelet coefficients, but we are trying to develop a non-stationary version of this filter, better adapted to the non-stationary characteristics of the financial market returns time series. Combined with the multivariate approach, this evidently implies a type of Kalman filter. Also, we would like to implement a display of a multiresolution analysis of the Wavelet spectrum for the security returns, based on the Maximal Overlap Discrete Wavelet Transform (MODWT).

QuanTek 3.3 (January 1, 2007)

In this new version, an attempt will be made to incorporate every existing technical indicator known to man, at least those that can be calculated using the available data and which have a chance of being meaningful. These indicators can then be tested for correlation with future returns and in simulated trading scenarios using the Diagnostic Test, just like the custom Momentum indicators based on the digital Savitzky-Golay Smoothing and Linear Prediction filters. These traditional indicators, or combinations of them, will also be added to the complement of Momentum indicators making up the Trading Rules. In this way it will be possible to test each indicator and verify its effectiveness in a variety of trading scenarios. These traditional technical indicators will also be available for display, either in the Main Graph or in splitter windows.

This update will take QuanTek to a new level of computational complexity. Some of the computations will be longer and require more memory than they have up to now, but these computations will be more elaborate tests of the Trading Rules and will not be required for everyday trading. A routine is now included to do the daily update of the Trading Rules without having to compute the entire set of Momentum indicators. I also have an idea for computing the optimal Momentum indicators automatically and eliminating the Lead Time adjustment for these indicators. Eliminating this adjustment, that might need to be tweaked continually, will make it possible to perform a completely "blind" test of the automatic trading rules in the Diagnostic Test, which will not depend on a manual setting of the Lead Time adjustment that was done "after the fact". This manual setting should still be useful, as it provides a way of optimizing the Trading Rules over the past data, in the hope that it will remain optimum for some time into the future, even though this manual setting cannot be tested directly for persistence into the future.

Some new statistical tests for serial correlation and heteroskedasticity will be added to the Correlation Test - Returns display.

We may also be able to upgrade the Wavelet LP filter and the optimal portfolio calculation.

QuanTek 3.2 (September 16, 2006)

This is (finally) the release version of QuanTek, after much testing and development of the Linear Prediction filter routines, in particular the development of a Wavelet LP filter. To go along with this, we have added a Hybrid LP Filter dialog, to enable setting the choice of Filter Type, Order of Approximation, and Fractal Dimension for each security data file individually. These filters and their parameters can be tested from within this dialog by calling the Correlation Test - Filters dialog, to measure the correlation between the raw Price Projection (returns) output of the LP filter with the future returns, for a variety of choices of the time horizon.

The Time Horizon adjustment can now be made separately for each security, rather than one setting for the whole program. This is important because each security has its own optimum time horizon for best performance of the Price Projection. Now this may be set either in the Correlation Test - Filters dialog when testing the filters or in the Correlation Test - Indicators dialog when testing the indicators.

A new dialog box, called the Trading Rules Filter & Momentum Weights dialog, was added, where the Trading Rules can be selected and displayed. The new selections for the Trading Rules include three new controls, called the Increment, the Threshold, and the Compression. In addition the three Momentum Weights, which are the weights of the three Momentum indicators in the Trading Rules, are set from this dialog. This new dialog is also called from the Diagnostic Test, so a set of Trading Rules based on the three saved Momentum indicators can be set and back-tested independently from the data file itself.

We have also implemented an upgraded version of the CrypKey licensing program, which will require all users to obtain new licenses from Omicron Research Institute. From now on two licenses will be supplied with each purchase of QuanTek. Also, a version of QuanTek for MetaStock has been developed, in addition to the version for TeleChart. These are separate "builds" of the same program, and use the same license, so you can download and use either (or both) interchangeably.

Version 3.2a (10/04/06): A problem with the normalization of the Momentum indicators and Trading Rules was fixed.

Version 3.2b (10/12/06): It was made possible to adjust the Time Horizon for each security individually. The Correct Harmonic Oscillator Phase dialog was added.

Version 3.2c (10/27/06): The Trading Rules Filter & Momentum Weights dialog was implemented in order to display the Trading Rules separately, and enable a wide range of settings of these based on the three saved Momentum indicators.

Version 3.2d (11/09/06): I have removed the Correct Harmonic Oscillator Phase dialog and demoted the Harmonic Oscillator indicator to display purposes only. Sorry about that. A 200-day simple moving average (SMA) has also been incorporated into the lowest scale of the Main Graph. A few miscellaneous bugs were also fixed.

Version 3.2e (11/20/06): I have removed the Price Projection from the Momentum indicators entirely. They did not appear to be adding much value, and were slowing down the calculations considerably. Also I felt it was important to filter all the Momentum indicator data, for each day, using the same filter, which is now just the Savitzky-Golay smoothing filter. Also the default Linear Prediction filter for the Main Graph was reset to the Fourier filter rather than the Wavelet filter. It still seems to be necessary to set the Fractal Dimension by hand for best performance, and the Fourier filter has the widest range of settings. Also a new set of Sample files was included.

QuanTek 3.1 (May 28, 2006)

This was the beta version of QuanTek, because the filter routines were still being developed and beta-tested. This version still only ran with TeleChart data, but for the next version, routines to read MetaStock files were being added.

This version was the first to incorporate the Wavelet Spectrum and the Wavelet Linear Prediction filters. These Wavelet LP filters were giving promising results, and we have expanded on this approach in subsequent versions of QuanTek.

Also a Correlation Test - Filters dialog was developed in order to test directly the predictive properties of the various Linear Prediction filters. In this version, six different LP filters were included, and any one of these filters can be selected to use for computing the Price Projection and Trading Rules. However, it was found necessary to be able to set the filter parameters manually, for each security individually, in particular the Fractal Dimension. This has been done in the subsequent version of QuanTek by bringing back the Hybrid LP Filter dialog, in addition to a choice of the six different filter types.

QuanTek 3.0 (January 16, 2006)

This first version of QuanTek was intended as a pre-beta version. This version ran with TC2000 v5.3 data files, and of course with the Sample files provided. The main purpose of this pre-beta version was to test it on the Sample files.

This version incorporated a Hybrid Filter dialog, in which you could design the Linear Prediction filter yourself, including the setting for the fractional difference parameter. This dialog was removed for subsequent versions, and replaced by a choice of six different fixed LP filters.

Features of QuanTek (2006)

QuanTek enables you to design your own custom technical indicators and trading rules, and to test these indicators and trading rules for effectiveness. The technical indicators are based upon a variety of Linear Prediction filters and the Savitzky-Golay smoothing filter. Used together, these two digital filters yield a wide variety of oscillator-type indicators. QuanTek also incorporates a variety of statistical tests to test the effectiveness of these indicators. There are several correlation tests to test the correlation between the indicators or the LP filter output with future returns directly. There is also a back-testing routine called the Diagnostic Test to test the Trading Rules derived from the indicators in several realistic trading scenarios with values of the time horizon for trading from 1 to 40 days. Also included are displays of the spectrum of returns, based on either the standard Periodogram or the Wavelet spectrum. These are included because these spectrum measurements are the basis of several of the Linear Prediction filters in QuanTek.

The QuanTek program also has a portfolio optimization routine, making use of the Markowitz Model to construct an optimal portfolio. You can use this optimization routine to adjust your portfolio to maximize return and minimize risk in the overall portfolio.

QuanTek now works with either TeleChart or MetaStock data, or directly with ASCII files in a variety of formats.

Here are some of the main features of QuanTek:

	Price Projection: A Price Projection is computed using one of a choice of Linear Prediction filters, and displayed on the Main Graph. This Price Projection is also used as part of the computation of the Momentum indicators and Trading Rules. One of the LP filters uses the Wavelet spectrum, one uses the Periodogram spectrum, and one computes the autocorrelation directly. At present, all of these filters assume a stationary stochastic process over the past 1024 days, which seems a reasonable approximation to the true non-stationary stochastic process underlying the financial markets. You can experiment with the results of using the different LP filters.
	Linear Prediction: The Hybrid LP Filter dialog is used to set the Linear Prediction filter for each security separately. The type of filter may be chosen, as just described, and two parameters which are called the Order of Approximation and the Fractal Dimension can be set. The Order of Approximation selects the degree of smoothing of the filter spectrum, by setting the length of the series of LP coefficients. The Fractal Dimension sets the low-frequency response of the filter, corresponding to modeling the time series as a fractionally differenced (FD) or *long-memory* process. The Hybrid LP Filter dialog also has attached to it a dialog box which displays the correlation between the filter output (returns) , and the future returns. Using this Correlation Test - Filters dialog, the predictive power of each filter can be tested directly for any settings of the filter type and parameters.
	Momentum Indicators: The main feature of *QuanTek* is that you can design and test yourself a wide variety of technical indicators based on the Linear Prediction filter and the Savitzky-Golay smoothing filter. These consist of practically every conceivable oscillator-type indicator that can be made out of the past price data. These indicators can then be tested for correlation with future returns using the statistical tests in *QuanTek. After a set of three of these indicators is designed, the Trading Rules* indicator is formed by a weighted sum of these three indicators, with various filter rules applied. The effectiveness of the final Trading Rules indicator can be tested in a variety of realistic trading scenarios by means of the Diagnostic Test.
	Statistical There are two correlation tests which measure correlation with future returns of the technical indicators and the output of the Linear Prediction filter. These give a direct indication of the effectiveness of these two methods of estimating future returns. There is also a Diagnostic Test, which is a back-testing routine to test the Trading Rules in a variety of realistic trading scenarios, for a given time horizon. Two of the statistical tests are the standard Periodogram spectrum and the Wavelet spectrum, which are two different methods for computing the power spectrum of the stock (log) price returns. These give another indication of the possible presence of correlation in the returns data.
	Main Graph: The scrolling graph has been improved, with four different scales that cover the whole data set. Now there is no limit to the time span of historical data that can be displayed. This also makes it possible to do a better statistical analysis, using deep historical data. The graphs are now available in either black or white background. The black background looks great! These four graph scales display the Price Projection, various buy/sell signals, Bollinger Bands, and on the highest scale is a Candlestick Graph.
	Portfolio Optimization: *QuanTek* also incorporates a portfolio optimization routine, making use of the Markowitz method. The correlation between the returns of all the securities in the portfolio (data files in a folder) is computed and used, along with the expected return for each security derived from the Price Projection, to compute the optimal portfolio which should maximize returns and minimize risk. The parameters for risk aversion and margin leverage can be set by the user, to achieve portfolios with different degrees of risk vs. return.

Features of StockEval (1998-2001)

StockEval was a preliminary version of our stock trading program, designed to incorporate up-to-date methods of computation and a scientific basis for technical analysis. This program featured the use of the Savitzky-Golay digital smoothing filter, as a replacement for moving averages, and a Linear Prediction filter to estimate future returns up to 100 days in the future. This resulted in a novel set of technical indicators which featured zero time lag, and a more straightforward interpretation than the usual ones. We experimented with several different approaches to Price Projection, each of which captures certain aspects of the overall problem, as it turns out.

StockEval used only data from Dial/Data or ASCII files. (It was written originally for Dow Jones data, and still had some legacy features from that, including the capability to log on to any database via modem as a Telnet program.)

These were the main features of StockEval:

	Price Projection: The main technical indicator in the StockEval program was a Price Projection, in which future prices are estimated up to 100 days in the future. Two of the main approaches we tried were a regression of the future returns on a set of exponential moving averages of the past prices, and the Linear Prediction routine. The first method used a set of exponentially weighted moving averages on different time scales, and the differences between EWMAs on adjacent time scales were taken as the basis functions (essentially MACD indicators). The future one-day returns were then regressed on these basis functions, and the whole procedure was extrapolated into the future one day at a time. This method apparently gave reasonable results. However, a simpler method is to utilize the standard Linear Prediction method, which is somewhat equivalent, although it leads to a "noisier" outcome due to a greater number of parameters being fitted. Both of these methods gave good results on the Diagnostic Test (a back-testing routine), but the excess noise in the trading rules on short time scales was a difficult problem. However, after applying the Savitzky-Golay digital smoothing filter to get rid of the short-term stochastic noise, the results were definitely positive, and led to rather spectacular short-term trading gains according to the Diagnostic Test, in the strongly trending market of the late '90s.
	Linear Prediction: The Linear Prediction filter used in StockEval was a standard, publicly available LP filter which models the time series as an auto-regressive (AR) process. This filter was used with the maximum number of parameters (1024) in order to try to get the best possible long-term prediction. (The short-term stochastic noise was then filtered out using the Savitzky-Golay smoothing filter.) However, after analyzing this particular filter, it turns out that it is actually adaptive in a certain sense (when used with the full set of parameters), in that it gives greater weight to the most recent data. This explains why we were getting pretty good results on the short-term trading rules as well, as measured in the Diagnostic Test.
	Harmonic Oscillator: The StockEval program featured a set of three novel technical indicators based on the Savitzky-Golay digital smoothing filter, which we call the Harmonic Oscillator indicators. The Savitzky-Golay filter can compute derivatives (rates of change) of the smoothed prices, and these smoothed prices and their first and second derivatives were used to construct the set of Harmonic Oscillator indicators with zero lag and a more straightforward interpretation than the normal ones. From these indicators, a set of trading rules was derived. These trading rules gave good results on the Diagnostic Test, although they only applied to a trading scenario in which the position is varied smoothly by making a small adjustment every day (which would only be practical for large portfolios).
	Main Graph: The Main Graph of StockEval was rather novel. It was a scrolling graph of the entire data set, presented in one long "panorama" which you could scroll across. There were four different scales on the graph, separated by a factor of two, and the horizontal and vertical axes both scaled the same way, so that the slope of the graph was preserved when rescaled. The price axis of the graph was logarithmic, so a constant distance along the vertical direction represented a constant percentage change. This made it possible to judge the relative performance of different stocks directly "by eye", just by comparing directly the slopes of the graphs (a feature not seen in the usual stock price graph). Underneath the log prices, the log volumes were plotted, relative to a reference line that represented the average log volume. This made it easier to see when the volume is larger or smaller than its average value.
	Portfolio Optimization: StockEval also incorporated a portfolio optimization routine, making use of the Markowitz method. The correlation between all the stocks in the portfolio (stock data files in a folder) was computed and used, along with the expected return derived from the Price Projection, to compute the optimal portfolio which would maximize returns and minimize risk. The parameters for risk aversion and margin leverage could be set by the user, to achieve portfolios with different degrees of risk vs. return.

As always, "Past performance is no guarantee of future results."

Please read and understand the Disclaimer.

Go back to the Home Page.

Last modified 11/25/2006 .