Question

Read the statements below. For each of the statements, select TRUE FALSE or NEI (Not Enough Information) accordingly. When evaluating different forecasting methods, the best option is always the one that gives you the least cumulative forecasted error: The CFE is useful in assessing bias in a forecast A value greater than 15000 for a Mean Absolute Deviation should be always considered as a large error: In the SMA, a shorter moving average period produces more oscillations but it more closely follows the actual trend.[ Using a WMA with n=4 will perform better than a SMA with n=4. For Exponential Smoothing, if actual demands are stable you should use a large a to lessen effects of short-term changes in demand.

Answer 1

It is true that the least value of CFE is preferable as the best way to improve forecast accuracy is to focus on reducing forecast error

It is true that CFE is useful in accessing bias in forecasting as it is defined as the measurement of total forecast error that assesses the bias in forecasting.

It is false that greater than 15000 MAD means larger error as MAD only tells about whether the mean is useful indicator of the values in the given data, it tells about the level of variation in the data.

It is true that a shorter-term moving average is more volatile, but its reading is closer to the source data.

It is false that WMA produces better results for shorter time period

It is true that if demand is stable then large value of alpha lessen the effect of short term changes in demand as large value of alpha will give the forecasted values close to last period’s data and hence more accurate forecasting.