|  DEC 1985

Part II. Implementation Using Alpha-Beta Filtering A trend following method for all seasons by Anthony W. Warren, Ph.D.

Part II. Implementation Using Alpha-Beta Filtering A trend following method for all seasons by Anthony W. Warren, Ph.D. Summary In the June issue of Technical Analysis of Stocks & Commodities we presented a trend following method for stock and commodity trading which has a number of advantages over moving averages for trade timing. The method consists of computing an upper and lower trend channel which defines an uncertainty band within which the market price fluctuates, and computing a trading filter which is used to make buy, sell, and hold decisions. Historical price data, the trend channels, and the trading filter are then plotted together to present a graphic analysis of the current trading environment. Figure 1 shows an example analysis using this method, where the price data consists of weekly closing prices of the Fidelity Equity mutual fund. The trend analysis plots are interpreted as follows: (1) If the trading filter lies below the lower trend channel, an uptrend market is in progress and one should buy or hold long market positions. (2) If the trading filter lies between the trend channels, no trend is evident and one should liquidate any current market positions. (3) If the trading filter lies above the upper trend channel, a downtrend market is in progress and one should sell or hold short market positions. With this method, trade timing decisions are based on penetrations of the lower and upper trend channels, i.e., one should buy or sell long positions based on penetration of the trading filter through the lower trend channel and should sell or cover short positions based on penetration of the upper trend channel. This method has two primary advantages over moving averages for trend following:

by Anthony W. Warren, Ph.D.