What is Expectation Modeling?

EM is an exciting mathematical breakthrough that addresses the fundamental question of whether the dynamic determining a pattern of equity returns will continue to remain intact. By mathematically modeling the growth of historic cumulative returns, it is possible to determine an accurate expectation of future risk-adjusted returns. EM uses the mathematical model of that expectation to quantify the probability of whether the collective dynamics contributing to growth in the equity curve will continue, or whether the pattern of returns has undergone a paradigm shift.

The algorithms and methodologies for EM are a proprietary invention of Trading Sciences. In particular, Ron Brown’s expertise with authoring modeling software for fitting curves, surfaces, overlapping peaks, and overlapping waveforms has been leveraged to produce something elegant and effective that answers this essential question. The specifics of the modeling remain part of Trading Sciences’ proprietary intellectual property.

Where does EM add value for an investor?

By modeling a security’s cumulative return, its growth can be separated from the mechanisms that produced the return. This means that EM produces a very useful metric for the entire class of instances that generate a pattern of cumulative return behavior.

To illustrate, long term holds of securities, futures, and ETFs, whether long or short, represent one extreme of where EM is useful. In these instances, EM is looking at the underlying dynamics that have resulted in the growth in the price of these entities, or their decline in the case of entities that have been sold short. In both directions, there is an expectation that the collective market mechanisms generating the wide-sense movement in price will continue into the future. EM is an effective tool in ascertaining when that assumption becomes invalid.

The equity curve of a fund that involves a basket of multiple securities that are frequently traded, including the actively traded scenario of an automated trading system, represents the other extreme where EM is valuable. In this case, EM offers an assessment that looks mostly at the trading system and the portfolio balancing that are the primary components of growth in equity. If the trading system becomes more or less effective, or the portfolio balance changes because of entities that behave in new or unexpected ways, EM quantifies the probability that something essential to that growth has changed.

Additionally, as an option trader, this information benefits you directly because it allows you to monitor the behavior of an underlying instrument and revisit your option strategies if need be.

How do I utilize this information?

Knowing that a risk-reward pattern has changed is one of the most important flags you can have as a trader. How many times have you heard an investor speak of how good they are in selecting securities, but that their main difficulty is determining the proper time to exit a position? This is an example of the simplest instance where EM is invaluable.

If the equity curve consists of a long-term buy and hold of a single entity, be it long or short, there is often an intrinsic assumption the dynamics generating the current growth or decay remain constant and the risk-reward likewise remains favorable. Knowing that such assumptions can no longer be deemed valid is immensely valuable in that it provides an early warning that such a position should be re-evaluated. For the fundamental analyst, there is a warning that something important has likely taken place in the market that suggests a re-evaluation of the risk-rewards expectation is warranted. Something fundamental has probably changed. EM detects both sides of such change. It not only quantifies those instances where the mechanisms for growth of the equity see unexpected decline, but also those where they see an unexpected rise. In the latter instance, the fundamental investor can look for the factor or factors that have produced the unexpected benefits.

If the cumulative curve analyzed by EM is generated by an actively traded system, then the elements or components of that trading are usually the primary factors EM addresses. The returns can derive from a trader’s active trades or the frequent trades generated by an automated trading system. The growth can be generated by any combination of long and short trades of one or more entities under investment. In this instance, EM warns that the underlying dynamics and mechanisms producing the growth in the investment have changed in such a way that one should review the trading strategy or trading system, or the entity or entities being traded. That is an immensely valuable early warning signal.

In this instance of an actively traded fund, EM furnishes a quantitative probability that a strategy, system, or portfolio has gone south and should be re-evaluated. That may mean it is time for a new strategy, for a revision to any trading system, or for a change of the components making up the portfolio. Here as well EM can also advise of positive changes in the collective paradigms generating the growth. Such information can be particularly useful in helping launch an effort to decipher what has changed, knowing that such factors may be invaluable going forward in managing trading strategies, optimizing automated trading systems, or selecting the securities that will be used in the basket of the fund.

What can you share about how EM works relative to similar tools?

We can share that the EM methodology models the risk adjusted growth in a manner somewhat similar to what one might see with Bollinger bands applied to a risk-adjusted equity curve, but without its critical flaws or weaknesses. If an equity curve falls outside the envelope of the Bollinger bands, it means that the performance or return has strayed a couple of standard deviations outside the growth or decay represented by a long-term moving average. That signals a fundamental shift has taken place, and a coarse probability that such has occurred can be derived from the distance the current adjusted equity is from the band.

Applying Bollinger bands or the CCI (Commodities Channel Index) type of analysis to a resultant equity curve are advanced techniques used by savvy fund managers and traders to ascertain whether their trading systems, techniques, or portfolios have undergone any kind of substantial change.

As valuable as such an analysis is, there are serious drawbacks. The central channel will contain a serious lag and the bands are based on deviations around such a lagged center. The estimation of these deviations contains additional lag as some window of time is needed to compute the estimate of scatter.

EM produces a central channel that contains no lag. The prediction limits from the mathematical model contain a true probability for the model as it is applied to the current day’s data. And quite importantly, EM incorporates proprietary technology to assess such on a fully risk-adjusted basis.

Does Trading Sciences employ EM for its own trading?

It has proven invaluable to us. EM was developed for our own trend/counter-trend and predictive trading systems. We use it extensively as our principal early warning tool that our trend system signals or predictive system signals are not working as expected. We use that warning to re-optimize our trend trading system and to rebuild the libraries of mathematical models for our predictive system.

Is EM available as a commercial software product for purchase?

No. Trading Sciences’ mission is to produce state-of-the-art leading-edge software technology for trading. The intellectual property in Trading Sciences’ technologies is available only as a solution that we offer to interested parties via licensing arrangements.

How do we follow up and determine whether EM represents a solution for our specific needs?

For more information on Trading Sciences, please email info@trading-sciences.com.