Trading IBM Intraday by Dennis Meyers, PhD
Here’s how to develop a trading system that measures the real price dynamics of the market.
The fading memory polynomial was first introduced in one of my previous articles, titled “The Yen Recursed.” In that article, I discussed how to use a first-order fading memory polynomial to trade the yen futures on a daily basis. Here, I will use a fourth-order fading polynomial to trade IBM one-minute bars on an intraday basis.
The fading memory polynomial is a mathematical technique that fits an nth-order polynomial to the last T price bars, but calculates the n coefficients of the polynomial such that the error between the polynomial and the current bar is weighted much higher than the error between the price and value of the polynomial n bars ago. As an example, if the latest price is at time t and the price made a turn at time bar t-10, then you do not want prices prior to t-10 affecting the polynomial fit as much. As is shown in the sidebar “Fading memory polynomial mathematics,” the most familiar case of the fading memory technique is the zeroth-order fading memory polynomial better known as the exponential moving average. The fading memory technique is in contrast to the least-squares polynomial fit, which equally weights all past errors between the polynomial and the price bar.