Cutting Edge Week 6 Forecast vs. Actual Daily Call Volume Chart: Day Last value

Question

Cutting Edge

Week 6 Forecast vs. Actual Daily Call Volume

Chart:

Day                 Last value        Averaging       Moving Average         Exp Sm 0.1     Exp Sm 0.5                

Monday           795                  982                  735                              802                  701

Tuesday           774                  946                  668                              768                  689

Wednesday      809                  1037                737                              837                  763

Thursday         947                  1227                833                              962                  773

Friday              759                  1032                676                              782                  572

MAD              171                  399                  104                              184                  116     

Actual Call Volume

Monday: 723

Tuesday: 698

Wednesday: 534

Thursday: 578

Friday: 697

*MAD=Mean Absolute Deviation

*Exp Sm= Exponential Smoothing

Answer Questions 2a through 2e below:

Question 2: Describe the details of each forecasting method used by Harry and explain its accuracy (MAD value) in comparison with the accuracy of the other methods. (Hint: In answering this question, it is helpful to review a time-series plot of the 13 weeks of data.)

2a) Last Value

2b) Averaging

2c) Moving Average (5 days)

2d) Exponential Smoothing (alpha= 0.1)

2e) Exponential Smoothing (alpha= 0.5)

Explanation / Answer

Ranking of methods in terms of accuracy:

1. Moving average

2.Exponential Smoothing (alpha= 0.5)

3. Last value

4. Exponential Smoothing (alpha= 0.1)

5. Averaging

2a) Last value method of forecasting is a naive method wherein the forecast of the next data point is the last value in time series. The sample size in this method is one i.e. last value. Given the MAD of 171, it ranks 3rd in the forecasting accuracy. In a volatile market (rapidly changing), this may be the only way to forecast accurately.

2b) This forecasting method takes into account all the data in the table and averages all data points to arrive at the forecast. It is the complete opposite of last value method wherein all data points are taken for the forecast. Since the call volume is highly variable, the averaging method is least accurate. The method may work better when the data points are more or less constant.

2c) The moving average method takes into account the recency and trend of data. The forecast is arrived by averaging data for last 5 days in this case and all data points have equal weight. The method has the lowest MAD and hence the most accurate.

2d) Exponential smoothing is a modified version of moving average. The most recent observation is assigned a weight and the older observations are assigned a smaller weight. The value of alpha depends on the process stability, A lower alpha signifies a stable process, however, there is a lot of variability and hence the method ranks 4 with 4th highest MAD.

The formula used = Forecast = Alpha* (Last Value) + (1-alpha) (Last Forecast)

5. Harry used a higher alpha which puts more weight to the recent values. Since the call volume is highly variable, the more weight on recent values gives a better MAD and the method ranks 2nd in overall forecasting technique.

Related Questions

Navigate

• Browse (All)
• Browse C
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.