Moving average versus seasonal indices on two years of frozen pierogi shipments: MT296's Unit 2 forecast exercise scores both on year two and explains where each goes wrong. Searches like "mt 296 unit 2 assignment example", "mt296 unit 2 sample" and "mt296 unit 2 example" land here.
What a finished MT296 Unit 2 demand forecast exercise looks like
About six pages, most of them tables. The data table lists 24 months of potato-cheddar cases, with the pattern plain to see: a Lent rise in February and March, a July low near 5,500, and a December peak above 12,800 both years. Method one is a three-month moving average. Method two computes a seasonal index for each month from year one, from 0.70 in July to 1.65 in December, then applies it to a deseasonalized three-month average. Both forecast year two month by month. The error table reports mean absolute deviation, 1,871 cases against 535, mean absolute percentage error, 22.8 against 6.6, and a tracking signal near zero for each. A final table isolates February through April, where the seasonal method's errors cluster.
How a MT296 Unit 2 example is structured
Every forecast in the exercise can be recomputed by hand from the data table, and the layout is chosen for that. Method descriptions come first, each with its formula and one worked month: the moving average for May as the mean of February through April, the seasonal forecast for the same month as that mean deseasonalized, averaged and multiplied back by May's index. The error section defines each measure before reporting it and explains why a tracking signal near zero shows neither method is biased overall. Interpretation follows. The moving average trails every turn in the season and misses December by 5,153 cases. Outside February through April the seasonal method averages under five percent error, and the paper traces its spring misses to Lent starting earlier in year two. The choice and its caveat close the paper.
Twenty-four months in one table
Every month's cases appear in a single table, letting any forecast be checked against the raw history without hunting through appendices.
Seasonal indices from year one
Each month's first-year volume divided by the first-year monthly mean gives indices from 0.70 in July to 1.65 in December, with February and March both above 1.2.
Three error measures, defined first
Mean absolute deviation in cases, mean absolute percentage error, and a tracking signal for bias. A one-line definition precedes each value, which keeps the comparison readable.
The moving average in December
Averaging September through November produces 8,037 cases for a month that shipped 13,190. That single miss illustrates why a lagging method fails a strongly seasonal product.
Lent moved, the index did not
Year two's February ran 1,248 cases above the seasonal forecast while March and April fell short, a pattern consistent with an earlier Ash Wednesday. Month-based indices cannot see a moving holiday.
Where marks go in MT296 Unit 2
Accurate-looking forecasts with no error measure beside them tend to draw the first comment from instructors, since the exercise asks for a comparison and a comparison needs a yardstick. Choosing a method by eye, the forecast line that looks closer, counts for less than choosing by computed error. A frequent technical slip is measuring error on the same months used to build the seasonal indices, which flatters the seasonal method; holding out year two avoids it. Papers that report percentage error but never ask whether the misses lean one way leave half the diagnosis undone. Explaining the largest misses in business terms, a holiday that moved or a promotion, earns more than simply listing them. The final choice should also admit what the winning method still cannot see.
Get a MT296 Unit 2 example written to your instructions
Attach the Unit 2 demand series, or describe the product if data must be found, together with the prompt, its rubric, and any method the instructions name. Two methods are run and compared on held-out months in the sample, all arithmetic shown. It comes free as a first custom sample, inside 24-48h.
MT296 Unit 2 questions, answered
Which two forecasting methods should I compare?
Whatever the prompt names comes first. If the choice is yours, pair a simple method with one that handles the data's main feature: a moving average against exponential smoothing for steady demand, or against a seasonal method when the series clearly repeats each year. Comparing two methods that share the same blind spot teaches very little.
What is a holdout period, and do I need one?
A holdout is data kept out of model building and used only to test forecasts. Building seasonal indices from year one and testing on year two is a simple version. Without one, error measures describe how well a method fits the past it was built on, which overstates how well it will forecast months it has never seen.
How are large forecast errors best explained?
Look for a business cause before a statistical one: a holiday that moved, a promotion, a stockout that capped sales, or a new customer. Name the likely cause, say how the method could be adjusted, and keep the adjustment out of the comparison itself so both methods are still tested on equal terms.