mean absolute percent error B u s i n e s s F i n a n c e

Attached file contains some of the data for “Performance Lawn Equipment” a company selling mowers and tractors.

1. Create trendlines for both Mower and Tractor sales for the column “World” in year 2014 only. Determine the best fitting trendlines and from the equation, predict the sales for mowers and tractors in April 2015. Hint: For determining X value for April 15, assign numbers 1,2,3,…. to time series (X) data starting from Jan 2014 all the way to April 2015.

2. Use the “Mower Unit Sales” “World” data for 2012, 2013 and 2014 and calculate forecasts for next 3 years (2015, 2016, 2017) using exponential smoothing method. Note: 2012-2014 data will be considered the actual values and you will calculate forecasts for 2015-2017 using those actuals. Assume the first period’s forecast for 2015 to be 7100 Units. For exponential forecasting, give α values from 0.1 to 0.9 and calculate 9 sets of forecasts. For each forecast, calculate MAPE (Mean absolute percent error) and determine which value of α you would use for your forecast based on your calculated MAPEs.

3. Use the “Tractor Unit Sales” “World” data for 2012, 2013 and 2014 and calculate forecasts for next 3 years (2015, 2016, 2017) using exponential smoothing with trend adjustments method. (We did not include this method for midterm exam but it’s widely used when there is presence of a trend in data which is the case for tractor sales). You can find the formulas on powerpoint page 37. Note: 2012-2014 data will be considered the actual values and you will calculate forecasts for 2015-2017 using those actuals. You will calculate Ft, Tt and FIT each as a column. FIT column will be the SUM of Ft and Tt in each row. Use α=0.4 and B=0.2. Assume the first period’s sales (Ft) to be 2500 Units and the first period’s trend (Tt) to be 100. Then Calculate MAD for your FIT forecast.