Technology Information

 


Frequently-Asked Questions

 

  1. What is a neural net?

A neural network is a mathematical modeling tool that has the capacity to learn by example. This is an extraordinarily useful ability, especially in financial modeling, where the there are usually countless examples. Neural networks use a different technique from standard analysis that is better suited to real world problems. 

  1. What is neural net training?

Networks first need to be trained by being presented with hundreds or even thousands of facts, each fact consisting of inputs and corresponding outputs. Through a unique feedback process, the network learns how those inputs are related to the outputs, and develops a general model that describes the relationship. 

  1. How do you know if training is successful?

If training is successful, then it will understand how to interpret new information.  In all neural net training runs, some data is held back for testing and additional validation.  If the net has not learned anything, and merely memorized its training data, then it will score well when presented with the training data, but poorly when presented with new fresh data.  A successfully trained net performs virtually the same on both sets.

  1. Why use a neural net instead of a generalized regression?

Mathematical models are normally built by making a priori assumptions about the functional form of the solution. These are called parametric models, and are solved by regression methods to determine a number of coefficients. This is sufficient if you know that the solution must be a second order polynomial, or some other simple, well-known function. But in the real world, relationships are not necessarily simple. Inputs and outputs could even be related in a non-linear fashion. If you do not have to guess the functional form of the answer, you have a big advantage.

  1. Why use a neural net instead of simple screening rules?

Lets say you have three indicators that are predictive of future price trends, A, B, and C.  Empirically you find that when A is up and B is down and C is up, that price tends to rise for several weeks.  One could set a threshold or make a rule using A, B, and C, but by doing this you might miss some signals.  For instance, what if each time A is up 50% more than B, the reading of C may need to be down to get good signals.  These trade-offs and higher order combinations are very hard to ferret out of the data.  Neural networks are designed to do this.  If a network were trained on many examples of A, B, and C, then only one rule would ever be necessary, such as “buy when the neural net reading is greater than 0.95.  All possible trade-offs between A, B, and C that result in an up forecast are being considered within the trained net.

  1. What are some of the inputs used in the Marque Millennium T Neural Net?

The MM  “T” neural net uses eight indicators that are quantitative in nature.  They measure behavioral patterns such as trend persistency, unusual volume activity, volume trends, price geometry, relative volatility, and non-linear cycles.  These indicators are proprietary, and were built using the principles of Chaos Theory.  The goal of our behavioral studies is to find early telltale signs of persistent positive investment behavior.  The most important manifestation of positive behavior is trend persistence, and the associated expansion of investor interest as price increases.  We use the Hurst exponent to measure trend persistency, and volume accumulation to measure investor interest.  Of course, the persistence of a trend is always threatened by the tendency for investors to become overly exuberant.  Every trend tends to go to an extreme, and evoke its own reversal.  Over indulgence and exuberance are shown by extreme accelerating trend persistency, with associated increases in volatility, and extreme turnover.  Our model is sensitive to these factors, and can anticipate reversals.  Trend lines and channels exist because investors make buy/sell decisions as price contact these invisible price lines, thus creating and reinforcing them through feedback. We map trend channels, because they provide feedback about performance expectations.  Within the geometrical context of these channels, our studies measure the rates of change and acceleration of trend persistency, volume accumulation, volatility, price disparity with respect to expectations, and behavioral time cycles. 

  1. What is Chaos Theory?

Chaos theory is the scientific theory that deals with non-linear systems.  These are systems in which some of the outputs are fed back in to form the next set of outputs.  Auction markets, while not being perfect non-linear systems, do demonstrate many of their characteristics.  Investors exhibit herd behavior on occasion.  By understanding these characteristics Marque Millennium has been able to build better indicators than has previously been possible.  

  1. What is the Marque Millennium T Neural Net trained to predict?

The “T” neural net was trained to find stocks that would outperform the S&P 500 index over a one-year holding period.

  1. How well does the Marque Millennium T Neural Net work by itself?

 

 

  1. What are some of the inputs used in the Marque Millennium F Neural Net?

The fifteen factors listed below are used to train a series of neural networks, which are then used to find the fair market value for a stock.

 

1        Earnings

2        Sales

3        Return on Equity

4        Growth in Sales

5        Adjusted Book Value

6        Cost of Capital

7        Industry Group (SIC)

8        Dividend Payout Ratio

9        Internal Growth

10    Operating Income

11    Economic Value Added

12    Operating Income as % of Net Sales

13    Net Recurring EPS as % Sales

14    Cost of Goods Sold as % Net Sales

15    Net Recurring EPS as % Operating Income

 

  1. What is the Marque Millennium F neural net trained to predict?

The Marque Millennium equity-pricing model uses neural networks to learn how a stock’s fundamental balance sheet data has been translated to a market price in the past, within the context of its industry group and economic sector. The models created for each industry group are then used to estimate current fair market values for these stocks.

  1. Why not just use valuation screens like other equity managers?

Equity screening has the drawback of not considering all fundamentals and their relationships at once.  If a P/E is too high, a stock might be thrown out, while its sales, book, and cash flow might be exceptional, and argue for its inclusion.  Our neural models will consider all factors simultaneously within the context of the industry group.

  1. How well does the Marque Millennium F neural net work by itself?

 


  1. How well do the T and F screens work together on different market caps?