Three forecasting methods, scored by MAPE, weighted error, bias and a running tracking signal against 26 weeks of grass seed shipments, in a composite MT438 analysis for Unit 5. Searches like "mt 438 unit 5 assignment example", "mt438 unit 5 sample" and "mt438 unit 5 example" land here.
What a finished MT438 Unit 5 forecast accuracy analysis looks like
Six pages carry the analysis, anchored by one table of 26 rows and a tracking signal chart. Actual shipments total 119,841 bags from January to June, rising from about 750 a week to a peak of 14,455 in the week of March 30. Method A repeats the same week of 2025, when spring came early. Method B scales a three-year seasonal profile to the annual plan. Method C is the consensus: B raised 12 percent, plus 900 bags a week through the peak. In the error table, mean absolute deviation runs 1,841, 362 and 850 bags a week; mean absolute percentage error 45.7, 7.8 and 13.6 percent; weighted absolute percentage error 39.9, 7.9 and 18.4; and bias plus 6.1, minus 0.2 and minus 18.2 percent.
How a MT438 Unit 5 example is structured
The error convention is stated in the first paragraph, actual minus forecast, so a negative bias means over-forecasting and every sign later in the paper reads the same way. Methods are described next in enough detail to reproduce, including where each forecast came from and when it was frozen. Four measures follow, each defined before it is used: mean absolute deviation for size in bags, MAPE for size relative to each week, weighted error for size relative to total volume, and bias for direction. A paragraph explains why MAPE and weighted error disagree, since one treats a January week and an April week alike. Tracking signals come next, week by week. Interpretation separates bias from timing, and the recommendation keeps method B with a rule for reviewing overrides.
Actual minus forecast
One sign convention governs the whole paper. Negative errors mean the forecast ran high, so the consensus's minus 18.2 percent bias reads as over-forecasting without translation.
A spring that came late
Method A borrows 2025's early season, forecasting peak volume two weeks before it arrived. Its 45.7 percent MAPE reflects timing more than scale.
Why MAPE flatters the consensus
Errors in the consensus gather in peak weeks, where volume is highest, so weighted error, 18.4 percent, exceeds MAPE, 13.6. MAPE gives a quiet January week the same vote as a busy April one.
Twenty-two weeks high
The consensus exceeded actual shipments in 22 weeks, and its four low weeks missed by under 40 bags each. Its tracking signal passes minus 4 in week 5 and ends at minus 25.7, a bias no single good week could offset.
A signal that recovers
Method A's tracking signal falls to minus 11.3 in week 12, then climbs to plus 4.0 as late-season demand outruns last year. The swing marks a timing miss rather than steady bias.
Keeping B, reviewing overrides
The statistical profile wins on every measure, with bias near zero and weighted error of 7.9 percent. Overrides remain allowed but must name a promotion or a new store count.
Where marks go in MT438 Unit 5
Instructors grading MT438 forecast comparisons look first for measures defined before they are used and a sign convention stated once. Papers reporting error without saying whether it is actual minus forecast or the reverse leave bias uninterpretable. A single accuracy measure is rarely enough: MAPE, weighted error and bias answer different questions, and a method can score well on one while failing another, as the consensus does here. Tracking signals should be read over time, not as one ending value, because a recovery after a timing miss means something different from steady drift. Stronger papers explain disagreements between measures rather than choosing the most flattering one. Recommendations that keep a method should say what evidence would prompt a review, so the choice can be revisited when conditions change.
Get a MT438 Unit 5 example written to your instructions
Send the demand history and forecasts in your Unit 5 case, the prompt and your rubric. A composite analysis comes back with its sign convention stated, each error measure defined before use, a week-by-week tracking signal and a recommendation saying what would trigger a review. That opening custom sample costs you nothing and needs 24-48h.
MT438 Unit 5 questions, answered
How do MAPE and weighted absolute percentage error differ?
MAPE averages each period's percentage error, so every period counts equally whatever its volume. Weighted error, often called WAPE, divides total absolute error by total actual volume, so high-volume periods count more. In seasonal data they can differ sharply. Reporting both, and explaining any gap, shows the grader you understand what each measure rewards.
How is a tracking signal calculated?
Divide the running sum of forecast errors by the mean absolute deviation to date. A forecast without bias keeps the signal near zero; persistent over- or under-forecasting pushes it away. Limits of plus or minus 4 are a common rule of thumb for flagging review. Read it period by period, because a signal that crosses a limit and returns tells a different story from one that keeps climbing.
Should I use MAPE when some periods have very low demand?
Use it with caution. When actual demand is small, even a modest miss becomes a huge percentage and can dominate the average. Many analysts add a weighted measure or exclude near-zero periods with a stated rule. Whatever you decide, say so explicitly, since graders tend to notice when a low-volume stretch drives an accuracy result.