Advantages and Disadvantages of the Moving Averages Method

Moving averages are a fairly straightforward and adaptable way to measure trends in a given time series data. If we want to predict the movement of some variable values over time, we must smooth out the irregular pattern in the variable’s old values first.

Moving averages are a technique that can be used to achieve this. It involves taking consecutive overlapping groups or parts of the time series and calculating their moving averages (arithmetic means). The practice of averaging eliminates extreme variations in the provided data. We now list out some of the merits and demerits of the method of moving averages.

Advantages of the Moving Averages Method:

1. When compared to the least squares method, this method is easier to learn and apply because it doesn’t call for any difficult mathematical calculations.
2. The main benefit of using a moving average is the chance it allows us to concentrate on long-term trend movements in a time series without being distracted by influences from short-term noise.
3. Unlike the freehand curve method, this method is subjectivity-free since the periodic motions in the data, not the researcher’s subjective judgment, define the period of the moving average.
4. In contrast to the least squares method of trend fitting, the moving average approach is highly versatile in that a few additional observations may be added to the provided data without changing the trend values already determined. Simply put, the inclusion of additional observations will increase the number of trend values in the conclusion.
5. By setting the moving average’s period to match the cyclic movement’s period in the supplied series, oscillatory movements can be totally removed. A good period selection can help lessen erratic variations.

Disadvantages of the Moving Averages Method:

1. We are unable to obtain the trend values for all of the provided observations, which is a clear drawback of the moving average method. Depending on the duration of the moving average, we must exclude the trend values for some observations at both extremes.
2. The moving average approach cannot be used to anticipate or predict future values, which is the primary goal of trend analysis.
3. Choosing the period of the moving average is difficult, especially when the time series does not show cycles that are regular. 
4. The trend values provided by the moving average approach are skewed when there is a non-linear trend.
5. The smoothing of variations improves as the number of periods averaged grows, but the approach becomes less responsive to actual changes in the data.
6. Generally speaking, moving averages do not correct for time series effects like seasonal variations.