Adding Calculations to the Cube - Expert Cube Development with SSAS Multidimensional Models - page 155

The final result of this formula is shown here:
Moving averages
Moving averages are another very common and useful kind of calculation. While
the calculations we've just seen compare the current period against a fixed point in
the past, moving averages are useful in determining the trend over a time range, for
example, showing the average sales of the last 12 months.
Depending on our requirements, we'll want to use different time ranges to calculate
the average. This is because moving averages are generally used to show long-term
trends by reducing seasonal, or random, variations in the data. For example, if we
are looking at the sales of ice cream then sales in summer will be much greater than
in winter, so we would want to look at an average over 12 months to account for
seasonal variations. On the other hand, if we were looking at sales of something like
bread where there was much less of a seasonal component, a three month moving
average might be more useful.