You run Colgate and sell toothpaste. You are trying to forecast demand for 2022, and you have sales data for the past 7 years, in $M:

2015: 55

2016: 45

2017: 100

2018: 50

2019: 100

2020: 150

2021: 200

First, generate 5 forecasts using the following methods: naive, simple mean method, 3 period moving average, 2 period weighed moving average (with weights of 0.8 and 0.2 for the most recent and second most recent period, respectively. The last two forecastingmethods are the exponential smoothing method, one with alpha = 0.5 and the other with alpha = 0.8. Anthony is in marketing, and he's very worried about being understocked, sohe picks his favorite forecasting method based on this worry. Bria is in investor relations, on the other-hand is really worried about big forecasting errors in either direction, and she picks her favorite forecasting method using her preferred metric. Both Anthony and Bria only use the last 4 periods worth of actuals and forecasts to generate errors to calculate their preferred forecasting accuracy metric. What is the absolute difference in Anthony's favorite forecasting method and Bria's favorite forecasting method?