CALL US (888) 422-0159
leadership blog

SeeAhead™ A MATHEMATICAL MODEL FOR FINANCIAL COACHING

May 9, 2025 | by | |
0 Likes

INTRODUCTION

SeeAhead™ is our mathematical model used to identify stock market movement early in the process. It is incorporated into the PerfectCoaches® app to help users actively manage their financial portfolio.  A mathematical model is a widely used research and development tool providing a simplified representation of a real-world system using mathematical equations and variables. It is a way to describe and understand complex situations by translating them into a quantifiable form, enabling analysis and prediction. Let’s begin with some background.

I entered graduate school equipped with sound quantitative skills acquired at Baltimore Polytechnic Institute. I decided to test the power of mathematical modelling as a research technique to explore Sigmund Freud’s theory of personality development, particularly his concept of identification. Identification is the process where individuals internalize characteristics or traits of another person and incorporate them into their own personality. This internalization can lead to adopting similar behaviors, values, or even becoming like the person they identify with. It’s a generalizable phenomenon. An adult can identify with an athlete, celebrity, political figure or even a social or political movement. However, Freud focused on identification with parents, emphasizing in his book The Ego and the Id (1923) that “the first identifications in earliest childhood will be profound and lasting.”

Freud’s theories were presented as discussions spanning multiple books and essays, making it difficult to define and measure the variables in a scientifically useful way. For my PhD dissertation, the capstone of my graduate study of social psychology, I tackled this important problem. I used data from a longitudinal survey of 1200 high school students and chose multi-variate regression analysis as a mathematical modelling approach to test key elements of Freud’s development theory. My dissertation concluded that that the child’s perception of a parent’s nurturance and power are the key forces. In other words, other people influence us because we perceive them as powerful or caring or both. How does one establish that scientifically?

Multi-variate regression applies Pearson’s product moment correlation to an entire system of variables. The coefficient of determination R2 (“R squared”) measures the proportion of variance in the dependent variable that’s explained by all the independent variables in the model. R2 is calculated by a lengthy series of equations using matrix algebra to define the regression model. Multiple regression R2 values range from 0 to 1 (0% to 100% of the variance explained). A higher R2 means the model explains more of the variation in the dependent variable. I tested my model under each type of parent-child relationship (female with mother, female with father, male with mother, male with father). The highest R2 =.26 was for female children identifying with their father, where my model was able to explain about a quarter of the variance, an acceptable result for basic social science. It would not be acceptable for financial prediction models, as will be discussed.

At my dissertation defense, a professor asked, “How could you explain more variance?” The gist of my answer was, “Sigmund Freud would argue that the model needs data on specific events in a child’s life that a survey can’t measure. He would be correct, but we can’t afford to psychoanalyze a thousand kids.” The professors smiled knowingly at my understated dig on the expensive and time-consuming method the great Sigmund Freud made famous. Soon, after almost two hours of equations and graphs, one of them said something like “good answers, solid research.“ My advisor took the cue, declared my presentation a success, and closed the meeting. Ironically, that brief exchange has implications for predicting stock market fluctuation, a modelling problem I am working on many years later.

I began with this example for three reasons. First, to illustrate the breadth of situations where mathematical models can be used; second, to establish R2 =.26 as an example of predictive power; third and most important, to emphasize that the power of your model can be increased when specific events are included as variables. All true, yet when I pulled my richly bound but almost never opened dissertation off the shelf to take this photo of the hand-drawn model, I had an  eerie feeling, thinking, “Wow, can it really be true that I am actually using this now?”

THE  SeeAhead™ MODEL

The modelling challenge my team and I are tackling now is this:  How can we help PerfectCoaches® users acquire the habit of actively managing their retirement investment portfolio? PerfectCoaches is an app based on my patent (US patent 19,249,212) for habit formation. Any habit can be learned using the app, plus the app identifies 44 habits that are particularly useful. They are divided into five skills: the daily life skills of Personal Excellence, Healthy Living, and Smart Money; and the business skills of Leadership and Customer Service. Each skill consists of specific habits.

In general, people don’t have well-defined habits for managing their money. The SmartMoney skill can change that. It is a soup-to-nuts prescription for sound financial planning, focused on the importance of saving. It consists of four habits a user can learn.
  1. The Planning habit says curb discretionary spending to save 20% of your income (the 50-30-20 rule) so you can take advantage of the fact that wealth doesn’t add up year after year, it multiplies (the rule of 72)
  2. The Daily Learning habit says learn one thing new about stocks, bonds, or real estate investing every day
  3. The Active Investment Management habit says don’t let your money sit; habits are shaped by cues in the environment, and the cues for this habit are text and email alerts prompting users to consider adjustments to their portfolio
  4. The Think Retirement habit says plan on using only part of your retirement funds each year (the 4% rule) to avoid running out of money

Although the Active Investment Management habit is just one of forty-four covered in the app, arguably it has the most immediate and tangible effects.  In order to make the alerts useful, they must be based on sound, timely predictions. How do we provide that portfolio management tool inside of an app? SeeAhead is the name we use for that tool.

SeeAhead is a probabilistic prediction tool. A good overview of that field can be found in  A State-of-the-Art Review of Probabilistic Portfolio Management for Future Stock Markets (Longsheng Cheng etal.,  Mathematics 2023, 11(5), 1148; https://doi.org/10.3390/math11051148). To quote the summary:
Portfolio management has long been one of the most significant challenges in large- and small-scale investments alike. The primary objective of portfolio management is to make investments with the most favorable rate of return and the lowest amount of risk. On the other hand, time series prediction has garnered significant attention in recent years for predicting the trend of stock prices in the future. The combination of these two approaches, i.e., predicting the future stock price and adopting portfolio management methods in the forecasted time series, has turned out to be a novel research line in the past few years. That is, to have a better understanding of the future, various researchers have attempted to predict the future behavior of stocks and subsequently implement portfolio management techniques on them. However, due to the uncertainty in predicting the future, the reliability of these methodologies is in question, and it is unclear to what extent their results can be relied upon. Therefore, probabilistic approaches have also entered the research arena, and attempts have been made to incorporate uncertainty into future forecasting and portfolio management. This issue has led to the development of probabilistic portfolio management for future data. This review paper begins with a discussion of various time-series prediction methods for stock market data. Next, a classification and evaluation of portfolio management approaches are provided. Afterwards, the Monte Carlo sampling method is discussed as the most prevalent technique for probabilistic analysis of stock market data. The probabilistic portfolio management method is applied to future Shanghai Stock Exchange data in the form of a case study to measure the applicability of this method to real-world projects. The results of this research can serve as a benchmark example for the analysis of other stock market data.

Although time-series data, probabilistic models, Monte Carlo simulations, and other tools are useful for understanding long-term trends in the market, we want two additional ingredients for PerfectCoaches. First, the app is designed to change habits, so we want to give users a cue—in the psychological sense— prompting them to take action. Second, there are many users in many different retirement investment plans, so we want the cue to be specific to a given user’s plan.

Because market timing requires taking action when appropriate, the SeeAhead model enables SmartMoney to provide alerts telling the user to consult their broker about asset re-allocation in their retirement investments. Alerts do not advocate specific trades or refer to specific stocks. Rather, they are designed to identify market movement early in the process. We call the algorithm SeeAhead because it can predict market movement days before it happens.

From a commercial point of view, two things make SeeAhead different. First, unlike the typical mathematical models discussed in the Cheng paper, SeeAhead  does not rely solely on time-series data. Rather, it attempts to quantify the potential impact of historical events (a binary variable) based on an analysis of what happened in the market when those events occurred. This recalls the point I made at my dissertation defense, namely, that Sigmund Freud wanted to know specifically what happened to the child before he would predict an outcome.

The second thing that sets this apart is that its predictions can be focused on one particular retirement plan. As a predictive model, the basic mathematics is always the same, but the predictions can differ based on the structure of any particular retirement plan the model is being applied to. Like the multi-variate regression technique I used in my dissertation, the mathematics is always the same but the particular calculations differ on a case-by-case basis.

The nuts and bolts of SeeAhead are even more complex than the mathematical model I used for my dissertation. Here are some of the most important features at an operational level.
  • It is based on calculations for each specific retirement fund.
  • It uses a VIX value based primarily on S&P 500 data.
  • It is a historic trend analysis of the movement of stocks and ETFS grouped by asset class: large cap, mid cap, small cap, and international.
  • Data is updated each day.
  • The tables are constructed in an SQL database.
  • The calculations are complex and involve pair-wise comparisons of assets and asset classes.
  • The goal, as in most financial models, is a very high level of measurable predictability.

SmartMoney and SmarterAllocations, the stand-alone spin-off of the PerfectCoaches app, were designed for investors not brokers. However, brokers can serve as FacilitatorCoaches within the PerfectCoaches business model (see the OurClub.tech website for details). The broker doesn’t have to do anything except offer PerfectCoaches  to their clients. Otherwise, our existing customer service process takes care of the rest.  For fiduciary brokers compensated based on the size of a client’s portfolio, it can help clients increase their wealth by improved market timing and just generally interacting with their broker more often. Brokers who are compensated for transactions can increase the volume of trades. The key to it all is seeing ahead. Thanks.

No Comments

Post A Comment