Question

When using Time Series forecasting, what is true about Double Exponential Smoothing?

Group of answer choices

The double exponential smoothing technique should not be used for predicting seasonal data.

Double Exponential Smoothing is able to predict both a level effect and a trend effect in the data. It is good for predicting cyclical data.

When Alpha (level) coefficient and Gamma (trend) coefficient are closer to .1 (rather than closer to .9), this tends to smooth the prediction.

All of the answers are true.

Answer 1

When using Time Series forecasting, what is true about Double Exponential Smoothing?

The double exponential smoothing technique should not be used for predicting seasonal data.

Reason- Double exponential smoothing of the raw series: Double exponential smoothing would not be suitable over the raw series. This method is not suitable because we cannot use for a series that has seasonality.

This method is deemed more reliable for analyzing data that shows a trend. In addition, this is a more complicated method which adds a second equation to the procedure:

bt = γ(St – St-1) + (1 – γ)bt-1

Where:

• γ is a constant that is chosen with reference to α. Like α it can be chosen through the Levenberg–Marquardt algorithm.

When Alpha (level) coefficient and Gamma (trend) coefficient are closer to .1 (rather than closer to .9), this tends to smooth the prediction.

Reason- Perhaps one of the most confusing aspects of exponential smoothing is the damping factor. Damping factors are used to smooth out the graph and take on a value between 0 and 1. Technically, the damping factor is 1 minus the alpha level (1 – α). So we need a smaller alpha levels (i.e. larger damping factors), smooths out the peaks and valleys more than larger alpha levels (smaller damping factors). Smaller damping factors also mean that your smoothed values are closer to the actual data points than larger damping factors.