|
Home Product Download News Faq Forums About Altreva Adaptive Modeler FAQ
General
Installation
Importing
quotes
Performance
Achieving
optimal results
Agent-based
model
Features
Integration GeneralWhat is Adaptive Modeler?Adaptive Modeler is a tool for creating agent-based market simulation models for price forecasting of real world stocks, ETFs or other market traded securities. How does Adaptive Modeler differ from other trading software?Most conventional trading software based on technical trading rules supports the user in finding or creating a (mostly static) rule-based trading strategy by optimizing and back-testing on historical data. If one searches long enough, this approach will always produce a trading strategy that seems highly profitable on historical data. This however doesn't mean that this strategy will also perform well in the future when price behavior may be different. The apparent past success of the strategy has in fact merely been caused by repeatedly optimizing and back-testing on the same historical data. This tends to lead to overfitting (or curve fitting) and is likely to produce trading rules that fail when exposed to new price data. More advanced software may provide adaptive trading rules that automatically adapt to price developments using neural networks, genetic algorithms or other techniques. However, one adaptive trading rule will still not be able to capture the complex price behavior of a financial market caused by the interaction of various heterogeneous investors, and this approach still carries the risk of overfitting. In financial markets no single trading rule continues to beat the market for any long period of time. Financial markets are constantly changing and new trading strategies come and go, affecting price behavior and each other’s returns. As the market evolves, trading strategies need to evolve as well in order to stay profitable. Instead of optimizing one or a few trading rules by back-testing them over and over on the same historical data, Adaptive Modeler lets a multitude of trading strategies compete and evolve on a virtual market in real time. This means that every historical price is only used once for "testing" the trading rules (as in the real world). This process is also said to be unoptimized and walk-forward. The overall behavior of the virtual market is the basis for trading signals. Though technical trading rules still form the basic building blocks, Adaptive Modeler automates the process of creating new trading rules to adapt to market changes and also diversifies the risk of a single trading rule by using many different trading rules simultaneously to generate trading signals. Is Adaptive Modeler free?A free Evaluation version is available and can be downloaded here (no registration required). The Evaluation version does not expire. A Registered (paid) version with more functionality is under construction. The Evaluation version has some functional limitations compared to the upcoming Registered version. The Evaluation version is meant to give traders, researchers and other interested people an opportunity to get introduced to and experiment with the application and its technology. How long can I use the free Evaluation version?The current Evaluation version does not expire (there is no limited trial period). Is there a more professional version?A Registered (paid) version with more functionality is currently under construction. Is Adaptive Modeler Open Source?No, Adaptive Modeler is not Open Source.
InstallationWhat do I need to run Adaptive Modeler?The minimum system requirements for running Adaptive
Modeler are: How do I install Adaptive Modeler?1. Download Adaptive Modeler from the download page. Note: Before installing Adaptive Modeler, the installer checks if Microsoft .Net 2.0 or higher is installed. If this is not the case, Microsoft .Net will be downloaded from Microsoft and installed first (after your consent). This requires an internet connection and may take a few minutes. Administrator privileges may be required to install Adaptive Modeler and/or Microsoft .Net. How do I upgrade from previous versions?If
you are a user of previous versions of Adaptive Modeler, please take note of
the following before installing the new version: No backwards compatibility with data files from versions lower
than 1.0 Uninstalling previous installations
Importing quotesHow are quotes being retrieved?Adaptive Modeler reads quotes from files. Therefore the user needs to have a file with historical quotes of the security to be modeled in one of the formats accepted by Adaptive Modeler. Also it is recommended to have quote retrieval software running that automatically receives quotes from a data provider, exports quote files in the required format and updates them with new quotes on time. See the User’s Guide for more information on quote retrieval. Does Adaptive Modeler support (real-time) online datafeeds?No, Adaptive Modeler only reads quotes from files. Most providers of quotes data provide software to write the received data to files in a format specified by the user. Can Adaptive Modeler read MetaStock quote files?Adaptive Modeler can only read MetaStock ASCII files, not the binary MetaStock files. For more information on how to import MetaStock ASCII files, see the User’s Guide. What quote data formats does Adaptive Modeler support?Adaptive Modeler accepts MetaStock ASCII files, Yahoo CSV files (some adjustments required), MetaTrader4 CSV files and other ASCII (CSV) formats. See the User’s Guide for more information on how to import these file formats. Can I use intraday quotes?Yes. See "What quote intervals are supported?". What quote intervals are supported?Adaptive Modeler supports any custom quote interval (intraday or end-of-day) with a minimum length of 1 second (probably even shorter) provided that quotes can be retrieved fast enough and that Adaptive Modeler’s processing time per quote (usually a fraction of a second) is less then the quote interval. Also quote intervals longer than one day are supported. How many historical quotes do I need?The more the better. Since Adaptive Modeler uses evolutionary computation (meaning that it learns over time) a sufficient number (at least a few thousand) of historical quotes of the selected security is needed. Only after a sufficient number of quotes has been processed, any statistical significance can be attributed to demonstrated forecasting success. How does Adaptive Modeler process splits and dividends?Adaptive Modeler does not automatically adjust for splits and
dividends. When will Adaptive Modeler retrieve a new quote?Adaptive Modeler retrieves new quotes as soon as the quote file has been updated (except when the model is paused). Why does a quote not get imported although it is in the quote file?When a quote does not get imported even when there is a valid and complete entry in the quote file, this can have the following causes: 1. The model may be paused. Press F3 to resume model evolution. 2. The quote file may (temporarily) not be accessible because another application is locking the file. Make sure that the quote file is not unnecessarily being locked by another application. How can I correct invalid or incomplete quotes in the quote file?When quote data is found to be invalid or incomplete by Adaptive Modeler, a warning is shown and model evolution is paused. The user can now correct the quote file (see the User’s Guide for quote data requirements). After correcting the quote file, the user should resume model evolution (press F3) and operation will resume normally. Erroneous quote data that might not be detected by Adaptive Modeler as invalid or incomplete, should be corrected before model evolution comes within 100 quotes of the quotes to be modified or else they may already have been read into a buffer and will not be read again. Adaptive Modeler can't read my quote file. What should I do?Make sure that the quote file is properly formatted in one of the supported quote file formats (see User’s Guide). If you still experience problems, please report them on the forum (Quotes section) and include a small part of the quote file.
PerformanceWhat returns can be made with Adaptive Modeler?The potential performance (in terms of risk and return) of trading based on a model's trading signals depends on:
The forecast success rate in its turn may depend on:
It is therefore not possible to say anything in general about the potential performance of Adaptive Modeler. Adaptive Modeler is a tool for creating market models that produce price forecasts. The accuracy of the forecasts of a specific model depends on all the factors given above. Adaptive Modeler is not an always winning trading system and its performance depends on how the user uses it. As an example of what is possible with Adaptive Modeler, an example model is provided on the download page. This is an actual model created with the Evaluation version that can be used by everybody for generating forecasts and trading signals into the future. Also, as its parameters are visible, it is easily possible for anyone to reproduce this model (by using the same model parameter values including the same random seed value). The example model serves as a proof of concept and is not necessarily the best model that can be created with Adaptive Modeler. However, we do not provide details about the performance of other models that we may have created for our own use or on behalf of others. As these models are proprietary, it would be inappropriate to provide unverifiable information about their perfomance. As a software publisher we wish not to make claims about investment returns. Also, such information is in fact largely irrelevant to users since they will not be able to use those particular models nor is there any reason to expect that those models are the best possible models that can be created with Adaptive Modeler. Obviously, there are many different combinations of securities, quote intervals and parameter values. The vast majority of these combinations have not been explored by us. We don't extensively explore the search space of combinations to prevent unintended optimization (overfitting) of the internal design of the agent-based model to historical market data. It is therefore likely that other combinations of security, quote interval and model parameters exist with better performance. Exploring the potential performance for a given security and quote interval requires experimentation and careful observation of results. With a given set of model parameters values, the results of different runs (separate model evolutions) can still vary because of random factors inherent to genetic programming. It may therefore be necessary to do a number of runs with the same parameter values for a reliable analysis of results. Adaptive Modeler provides a variety of ways to review the quality of previous forecasts and trading signals. Adaptive Modeler contains a Trading Simulator with a Performane Overview showing various return/risk indicators and sub period statistics. Also it is possible to project likely future trading returns under given conditions through stochastic simulation. However, as with any system that aims to make predictions about the future, there is no guarantee that any demonstrated forecasting success or trading performance will remain the same in the future. The user should be aware of this and consider the risks and potential rewards of every investment or trading decision on its own merits. How can I see the performance of the Trading Simulator?Open the Performance Window through the "View" menu. (If it is already open as an unselected tab, then click on the "Performance" tab to select it). The Performance Window should now be showing. A "Settings" button in the top-left corner allows changing the calculation period and other settings. For more information, see the User’s Guide. Can Adaptive Modeler predict the next market crash or bubble?Adaptive Modeler only forecasts one bar ahead. This way the most recent market price data is always available to the agent-based model when calculating a forecast. Therefore, a market crash or bubble can only be anticipated bar by bar as it unfolds. The effect of forecasting multiple bars ahead could be imitated by evolving a separate model using a longer quote interval. Why are the results of different runs different?The results of different runs (separate model evolutions using the same quote data and model parameters) can still vary because of random factors inherent to genetic programming. Random factors are used in the creation and evolution of trading rules. (Note that to recreate a model exactly it is necessary to use the same quote data, the same model parameter values, the same random seed value and the same version of Adaptive Modeler). For a more complete analysis of the potential performance of a given combination of quote data and model parameters, it is recommended to do a number of runs to see the variation in performance indicators. For more information about the creation of trading rules through genetic programming and running multiple model evolutions, see the User’s Guide.
Achieving optimal resultsWhat securities and quote intervals does Adaptive Modeler work best on?This has not been extensively researched by us. Many different stocks, ETFs, forex currency pairs, commodities and other securities are being traded on financial markets around the globe. Also, several different quote intervals could be used for each of these securities. Different securities and quote intervals may require different model parameter values. We don't extensively explore the search space of combinations of security, quote interval and model parameters to prevent unintended optimization (overfitting) of the internal design of the agent-based model to historical market data. Adaptive Modeler is specifically designed for active trading of highly liquid large cap securities with small spreads, low broker commissions and sufficient volatility. These are typically large cap stocks or Exchange Traded Funds (ETFs). However any stock, fund, FX currency or other security that meets these criteria will equally qualify. Although the system can in principle process any kind of time series, it is important (in general) when selecting a security to consider the relationship between the average transactions costs (broker commissions, spread and slippage), the available capital, and the (expected) volatility on the interval one intends to trade. When these are out of balance it will be very hard to make a profit even with high forecasting success. In general it will be more difficult to achieve good performance on short intraday intervals than on end-of-day data because the volatility on short intraday intervals is usually relatively low compared to the transaction costs. This means that the break-even forecast success rate is higher than for end-of-day data. What parameters values should I use? What are the best settings?There
is no general answer to this. Some parameters clearly depend on the quote data
being used. For other parameters the default settings are a good starting point
but experimentation is strongly recommended. No person or team alone can ever
explore all the possible parameter value combinations for all the quote data of
different securities and intervals. As said earlier, we don't extensively
explore the search space of all possible parameter values to prevent unintended
optimization (overfitting) of the internal design of the agent-based model to
historical market data. Therefore the default parameter values are unlikely to
be the best values for all cases. How high should Forecast Directional Accuracy (FDA) be for a model to be profitable?This depends on factors such as Forecast Directional Significance (FDS), volatility, transaction costs and other Trading System parameters. In general, it should at least be above 50%. Then the forecasted direction of price change per bar is more often right than wrong. With the Statistical Simulation data series it is possible to project the potential returns based on expected values for FDA, FDS, volatility, transaction cost and other Trading System parameters. See the User’s Guide sections about Statistical Simulations for more information about this. Can trading signals automatically be enabled/disabled based on FDA?Yes. On the Trading System tab in the "New/Edit Model" dialog, check "Apply FDA filter" and enter the desired threshold value. The FDA calculation settings can be changed by clicking on the "FDA settings..." button. For how long are the trading signals valid?A trading signals remains valid until a new signal is given. For more details on how trading signals are being generated, see the Trading Signal Generator section in the User’s Guide. I can not seem to create profitable models on minute data. What is wrong?If your models are not profitable even though the Forecast Directional Accuracy (FDA) is clearly above 50% on average, then the bar-to-bar price changes are probably too small to cover transaction costs (broker commissions, spread and slippage). For small price changes, FDA needs to be higher to reach break-even than for bigger price changes. To increase FDA, try experimenting with other model parameters. If this still doesn't work, then consider using a longer quote interval. Which forecast is better: the Virtual Market Price or the Best Agents Price?By default, Adaptive Modeler uses the Virtual Market Price as the forecast. Optionally, the forecast can also be based on the Best Agents Price. This feature is useful since it allows a comparison between the predictive abilities of the Virtual Market Price (which is based on the behavior of all agents) versus that of the Best Agents Price (which is based only on a group of the best performing agents). Since an essential principle of Adaptive Modeler is to use the Virtual Market Price as the forecast, it may be interesting to see whether or not this in fact outperforms a forecast based on the behavior of only the best performing agents (which may seem more intuitive to some and more in line with methods generally used by trading software). As far as we have observed, the Virtual Market Price almost always performs better than the Best Agents Price. However, this may not be the case in all situations. Experimenting is recommended. Note that the accuracy of both forecasts can easily be compared by showing two FDA data series together in one chart and setting the source parameter of one data series to the Virtual Market Price and the other to the Best Agents Price. Also note that the Best Agents group size can be changed on the Model tab of the "New/Edit Model" dialog.
Agent-based modelHow many agents can a model contain?In the Evaluation version the maximum population size is limited to 2000 agents. How do agents make trading decisions? What kind of trading strategies do they use?Each agent has its own (technical) trading rule. The trading rules can use historical price and volume data as input and return an “advice” consisting of a desired position to hold in the security and an order limit price for buying or selling the security. The internal logic of the trading rules consists of various functions such as price and volume data access functions; average, min, and max functions; logical and comparison operators; and some basic Technical Indicators. In most cases, agents are technical traders. The trading rules are created and evolve through a special adaptive form of genetic programming. For more information about this, see the User’s Guide. It is also possible to simulate "zero-intelligence" trading by using the RndPos and RndLim genes. With these genes the position advice and order limit price are established randomly. This could be interesting for comparison purposes. (To simulate complete zero-intelligence trading, select only the genes "advice", "RndPos" and "RndLim"; disable "Create unique genomes"; disable breeding by setting "Breeding cycle frequency" to 100,000; set "Broker Commission for agents" to zero). Do agents learn?Through an evolutionary breeding process, agents with poor performance are regularly being replaced by new offspring agents. The trading rules of these new agents are being created through crossover of the trading rules of the best performing agents. This way (parts of) trading rules that perform well get copied and recombined while poor performing trading rules are being removed. This way, the population of trading rules as a whole attempts to adapt to changing market behavior. So the population is adapting rather than learning, as market behavior is dynamic. Technically speaking, agents themselves don't adapt or learn since their trading rule doesn't change during their lifetime. (Trading rules only "change" by the replacement of old agents and their trading rules by new agents with new trading rules). Do agents use "swarming" or other means of direct communication?Adaptive Modeler currently does not use swarming or any other direct form of communication between agents. This is still being researched. Of course agents indirectly exchange information through the Virtual Market. Are there different types of agents?No different types of agents have explicitly been defined in Adaptive Modeler. So there are no specific agents defined for giving buy or sell signals, no broker or market maker agents, no long term or short term agents, etc. However, agents all have their own trading rule directing their trading behavior in different ways. Therefore, groups of different types of agents (in terms of their trading behavior) may emerge through evolution. This can be observed with agent distribution data series that reveal trading style characteristics such as Trade Duration Distribution, Volatility Distribution and Beta Distribution (either in distribution charts or in the Population window). For more details, see the User’s Guide. How does market making take place on the Virtual Market?The Virtual Market is a simulated double auction market where all buy and sell orders from agents are collected. Every bar, after all agents have evaluated their trading rule and placed their order, the clearing price is calculated. The clearing price is the price at which the highest trading volume can be matched. All matching orders are then executed at the clearing price. For more details, see the User’s Guide. Is price impact take into account?Yes. In the Agent-based model, the Virtual Market clearing prices are based on the order limit prices of the agents. Because of the volume weighted clearing price calculation mechanism, an agent offering a higher bid price increases the chance of its buy order being executed and thereby having an increasing effect on the clearing price (vice versa for sell orders). Price impact may also be an issue in the Trading Simulator when (simulating) trading by amounts that may affect the real world market price of thinly traded securities. In this case price impact can be taken into account with the slippage parameter (on the Trading System tab of the "New/Edit Model" dialog). How are the forecasts and trading signals made?Adaptive Modeler calculates a new forecast every bar. The forecast is normally based on the Virtual Market price. This is the clearing price of the Virtual Market calculated using the volume weighted pricing mechanism that includes all agent orders. The forecast is therefore based on the buy and sell orders of a large number of agents. As explained elsewhere, the forecast can alternatively be based on the Best Agents Price. After every new forecast the Trading System determines if a new trading signal (long, short or cash) needs to be given based on the forecast, the last known security price and the Trading System parameters. A new trading signal will only be generated when the new suggested position differs from the last generated signal. For more details on how trading signals are being generated, see the Trading Signal Generator section in the User’s Guide. Does a model keep evolving when I add new quotes to the quote file?Yes. There is no difference in the way historical quotes and new quotes are being processed. The model keeps evolving with every new quote. Once a model has gone through the quote file, how can I use it for forecasting?When a model has reached the end of the quote file, the forecast for the next bar has already been calculated. The forecast is shown in the Current Values window (if not, drag the Forecast dataseries from the Data Series window into the Current Values window). The trading signals can be seen in the Trading Signals window. The most recent signal is at the bottom of the list. When a new quote is added to the quote file, it will automatically be read by Adaptive Modeler. The model will then evolve one step further and a new forecast (and trading signal if necessary) will be calculated. Is there a difference between a "training" phase and an "operational" phase?No. In fact, there are no "training" or "operational" phases in Adaptive Modeler. There is no difference in the way the model processes historical quotes and the way it processes new quotes. The model keeps evolving with every new quote that is added to the quote file. A model does not first repeatedly train or optimize on historical data. Every quote bar is experienced (traded on) only once by the agents. Since it is possible to switch the Trading Simulator on and off, it could be said that an "operational" or "trading" phase starts when the Trading Simulator gets started. However, this has nothing to do with the evolution of the Agent-based model or the calculation of forecasts. The Trading Simulator and the generation of trading signals has no effect on the Agent-based model.
FeaturesCan I edit an agent's trading rule?It is not possible to directly edit an agent's trading rule. It is however possible to influence the creation of trading rules through genetic programming by specifying what functions are to be included/excluded and by changing size and depth limitations of the trading rules. To do this, go to the "Genomes" tab of the "New/Edit Model" dialog. Can I provide agents with my own trading algorithms or systems through an API or scripting language?This is not possible. If agents would be able to use trading logic written in an external (unknown) language, genetic operators such as crossover and mutation could not be applied anymore. Also the control and calculation of genome statistics would become problematic. Can I change the selection criteria for breeding and replacement of agents?No, these selection criteria can not be changed. For more information on how the selection criteria (the Breeding Fitness Return and the Replacement Fitness Return) are calculated, see the User’s Guide. Can I use multiple input time series (multivariate approach)?This is not yet possible but being considered for future versions. Can I forecast multiple securities in one model?Adaptive Modeler currently creates "single security" models. So only one security can be forecasted by a particular model. Can I automatically create models for several securities and report the results?Yes. This is possible through the Batch function (from the "Tools" menu). Does Adaptive Modeler support options and futures?Any quote data that meets Adaptive Modeler's quote data format requirements can be fed into Adaptive Modeler but no specific functionality is included for futures, options or other derivative instruments. Is multibar or multi-timeframe forecasting possible?Adaptive Modeler forecasts one bar ahead. There is always only one forecast, and that forecast is intended for the next bar. The effect of forecasting multiple bars ahead could be imitated by creating a separate model that uses a longer quote interval.
IntegrationCan I export data from Adaptive Modeler to other programs?Yes. All data series (including forecasts and trading signals) can be exported to a CSV file (comma separated values) using the Export function from the "Tools" menu. Can I export data to Excel?After exporting data to a CSV file, the CSV file can be opened or imported with Microsoft Excel or any other program that can read CSV files. Does Adaptive Modeler connect with databases?No, Adaptive Modeler does not connect with databases. Does Adaptive Modeler support order routing to online brokers?Adaptive Modeler does not support automatic order placement with online brokers. Can Adaptive Modeler be integrated into MetaTrader4?No, this is not possible. However, it is possible to import .CSV quote files from MetaTrader4 into Adaptive Modeler. Is there an API or DLL so that programmers can use this technology in their own programs?No, this is not provided.
Contact Us Hosted by Bluehost.com |
|
|