Mean Absolute Deviation and Forecasting

The video tutorial explains the concept of Mean Absolute Deviation (MAD) and its use in evaluating forecast effectiveness. It covers the steps to calculate deviation from demand, the importance of absolute deviation, and how to compute total error. The tutorial also demonstrates how to calculate MAD and its application in comparing different forecasting models, emphasizing that a lower MAD value indicates a better predictive model.

What is the primary purpose of using Mean Absolute Deviation (MAD)?

• To find the lowest forecast value

• To determine the highest demand period

• To evaluate forecast effectiveness

• To calculate the average demand

What is the first step in calculating MAD after forecasting data?

• Calculating the total error

• Finding the absolute value

• Determining the sample size

• Calculating deviations

What does a high negative deviation indicate in forecasting?

• Very low forecasted value

• Accurate forecast

• Very high forecasted value

• No deviation

Why is it important to find the absolute value of deviations?

• To ensure deviations do not cancel each other out

• To reduce the sample size

• To increase the total error

• To simplify calculations

Which Excel function can be used to calculate the absolute value of deviations?

• AVERAGE

• ABS

• COUNT

• SUM

What is the result of calculating the total error without canceling positive and negative values?

• No change in total error

• A lower total error

• A higher total error

• A negative total error

How can you determine the sample size if it is unknown?

• By using the SUM function

• By using the MAX function

• By using the COUNT function

• By using the AVERAGE function

What does a lower MAD value indicate when comparing forecasting models?

• A less accurate model

• No difference in accuracy

• A more accurate model

• A higher demand

Why is it not sufficient to rely on a single MAD value for one model?

• It does not provide a complete picture

• It is too complex to calculate

• It is not used in forecasting

• It always results in zero