Custom Search

Moving Average Basics

Of all systems available to traders in the futures and stock markets, the vast majority are based on moving averages, or on a variation of the moving average theme. Moving averages are not difficult to underВэstand, relatively simple to apply, and frequently quite easy to calculate. There are many different types of moving averages. Here is a brief list of the moving averages I've worked with, followed by a brief definition:

1.ВрВр Simple Moving Average (MA).ВрВр This is a moving average in which each price of the data series is assigned the same weight or value.

A 10-period moving average is calculated simply by summing all 10 valВэues and dividing by 10. The second moving average value is calculated by dropping the first raw value from the sum and adding the eleventh value to the sum. This leaves a window of 10 data points which are then added and divided by 10. The result is a second value in the moving average series. The process continues with the next raw data point. A moving average of x time units in length always has x data points. When I refer to a 10-period moving average, I mean specifically 10 days, 10 hours, 10 years, 10 months, or 10 segments of 5 minutes each. The time length of the unit is referred to as the period.

Typically in our work with intraday moving averages, we will be dealing with moving averages of from one minute to one hour in length as the period. Consequently, if I refer to a 10-period simple moving average on the one minute data, I am referring to a moving average calВэculated by adding together the most recent ten one minute prices and dividing by ten. If I refer to a ten hour moving average I am referring to the last ten hours worth of prices (one price per hour), added together, and divided by ten.

2. Exponential Moving Average (EMA). An exponential moving average is calculated slightly differently than is a simple moving averВэage inasmuch as it exponentially weights each value. The purpose of using an exponential moving average is, theoretically, to provide an average which will be more responsive, theoretically, to the underlying data. There are numerous sources which you may consult for the specifics of exponential MA calculation.

3. Weighted Moving Average (WMA). This is one of my favorite moving averages, because it does not assign equal weight to all values in the data series. There are two types of weighted moving averages, front-weighted (also called front-loaded), and back-weighted (also called back-loaded). If we refer to the front portion of the data as the most recent data, then a front-weighted moving average multiplies each value toward the most recent data by a constant weight or value in order to provide for greater impact of current data. A back-weighted moving average weights the earliest data in the price series in order to give it more significance in the final analysis.

4. Smoothed Moving Average (SMA). A smoothed moving averВэage is a weighted moving average of a different type. In this case, the data is "smoothed" mathematically to keep the averages from moving around too much. This, it is hoped, will yield more stable signals.

5. Triangular Moving Average (TMA). Yet another type of movВэing average, the TMA is weighted to emphasize the normal statistical distribution. In other words, the extremes of the TMA are less heavily weighted than is the central portion of the TMA. This results in the TMA being more centered and, therefore, more responsive to the normal distribution qualities of the data series. In order to calculate a 7-day TMA, for example, the following procedure would be used:

Raw data: A, B, C, D, E, F, G (7 days) Step 1:

x = (1 x A) + (2 X B) + (3 x C) + (4 x D) + (3 X E) + (2 x F) + (1 X G)

Step 2: TMA = x/16 (16 = sum of all multipliers)