BAR · Business Analysis · 6 practice questions
Time-Series Forecasting: Match the Method to the Pattern
Choose a forecasting method that matches whether demand is level, trending, or seasonal. Below, compare the same distributor's demand with only the pattern changed.
Try one first
Hint
Match the method to the series pattern: level only, level plus trend, or level plus seasonality.
Answer C. Simple exponential smoothing is designed for a level series with random variation but no systematic trend or seasonality. The demand pattern described is stable around a constant mean, so a level-only method is the best match.
Why not A: This choice comes from assuming every time series should include a trend component. Trend-adjusted smoothing is appropriate when the series has a sustained upward or downward movement, which these facts do not show.
Why not B: This choice comes from preferring a more complex model by default. A regression forecast is generally used when demand is related to predictor variables, but no explanatory variable is part of this forecast problem.
Why not D: This choice comes from treating monthly data as automatically seasonal. Seasonal smoothing is used only when repeating within-year patterns appear, and the facts specifically say they do not.
Side by side
| Level demand | Trending demand | Seasonal demand | |
|---|---|---|---|
| Shared facts | A distributor reviews 30 months of monthly demand for one replacement part. | The same distributor, replacement part, and 30-month observation history. | The same distributor, replacement part, and 30-month observation history. |
| Demand pattern | Demand fluctuates around 1,000 units, with no sustained direction or recurring seasonal pattern. | Demand has the same underlying level but now rises by a steady amount each month, with no recurring seasonal pattern. | Demand has the same underlying level and no sustained trend, but December repeatedly runs 20% above its deseasonalized level. |
| Best method | Simple exponential smoothing. | Linear trend regression. | A seasonal-index model. |
| Tempting wrong choice | Seasonal smoothing merely because the observations are monthly. | A weighted moving average that reacts faster but does not directly model trend. | Simple exponential smoothing that ignores December's recurring increase. |
| Decisive fact | Variation has no sustained direction or recurring seasonal timing. | Demand rises consistently over time. | The increase recurs in the same month. |
Key points
- Remember: The demand pattern selects the method, not the fact that observations are monthly.
- The distributor, product, observation history, and underlying demand level stay the same; only the pattern changes.
- A more complex model is not automatically a better fit.
How the exam traps you
- Treating monthly observations as proof of seasonality. Look for a pattern that repeats in the same months.
- Choosing a weighted moving average because recent demand is rising. A sustained rise calls for a method that directly models trend.
- Adding trend or seasonal components to stable demand. Use simple exponential smoothing when neither pattern appears.
Question 2
Hint
Focus on what happens to the planning model after an actual period is completed.
Answer B. A rolling forecast is refreshed as actual periods are completed and extends the model by adding a later period. That differs from merely revising estimates within the original annual plan. A static budget generally remains fixed for the original period and serves as a variance benchmark.
Why not A: This mistake treats planning as an annual replacement cycle. It fails because a rolling forecast is maintained during the year rather than waiting for the next annual budgeting cycle.
Why not C: This mistake focuses only on incorporating completed-period actual results. It fails because a rolling forecast concerns maintaining future planning information, including adding a new future period.
Why not D: This mistake treats a rolling forecast as only a reforecast of the remaining original fiscal year. It fails because the rolling approach also extends the model by adding a later forecast period.
Question 3
Hint
Decide whether the number given is a base amount with seasonality removed or an actual amount that still includes seasonality.
Answer A. Under a multiplicative seasonal model, a deseasonalized forecast is converted to the period's expected actual amount by multiplying by the seasonal index. Here, $520,000 × 1.20 = $624,000. Because December's index is above 1.00, the forecasted actual revenue should be higher than the base level.
Why not B: This comes from dividing by 1.20, which is how seasonality is removed from an actual amount. The amount given in the stem is already deseasonalized, so division moves in the wrong direction.
Why not C: This reflects ignoring the seasonal index entirely. A December index of 1.20 means the forecast should be adjusted upward from the deseasonalized level.
Why not D: This results from subtracting 20% from the base forecast. In a multiplicative model, an index above 1.00 increases the forecast through multiplication, rather than reducing it.
Question 4
Hint
Look for a systematic error pattern tied to the pattern the model assumes will continue.
Answer D. A consistent sign pattern by season indicates systematic bias rather than random error. Negative summer errors and positive winter errors show the model is overforecasting one season and underforecasting another. That is evidence the historical seasonal relationship may no longer apply.
Why not A: This mistake treats large absolute errors as proof that seasonality failed. Larger errors in a high-volume season may occur even when the seasonal pattern is still valid.
Why not B: This mistake treats any error movement as a failed assumption. Alternating signs within a season are more consistent with random forecast error than with a changed seasonal pattern.
Why not C: This mistake treats a single tolerance breach as an assumption failure. A one-month error that reverses in the next month does not show a sustained seasonal shift.
Question 5
Hint
Match each period to its stated weight by recency, then add the weighted amounts.
Answer D. A weighted moving average multiplies each month's sales by its assigned weight and then sums the results. Using the stated order, June's forecast is (240 × 0.50) + (210 × 0.30) + (180 × 0.20) = 219 units.
Why not A: This results from reversing the weights and placing 0.50 on March and 0.20 on May. It fails because the largest weight must be applied to the most recent month, May.
Why not B: This comes from taking a simple average of the three months: (180 + 210 + 240) ÷ 3. It fails because the method assigns different weights, not equal weights, to each month.
Why not C: This comes from switching the 0.30 and 0.20 weights between April and March. It fails because April is the next most recent month and should receive 0.30, while March should receive 0.20.
Question 6
Hint
Use the period number defined by the equation, then apply the slope to that time index.
Answer C. A trend projection uses the equation's time index for the forecast period. Year 6 means t = 6. Substituting gives 1,200 + 75(6) = 1,650.
Why not A: This comes from using t = 5, often by counting only the years after Year 1. That fails because the equation already assigns Year 6 directly to t = 6.
Why not B: This comes from adding one period's increase to the intercept, as if the forecast were for Year 1. That fails because the question asks for Year 6, so the equation requires t = 6.
Why not D: This comes from using t = 7, as if the series began with Year 0. That fails because the stem states the count starts with Year 1 at t = 1.
Common questions
When should I use simple exponential smoothing?
Use it when demand fluctuates around a stable level without a sustained trend or recurring seasonal pattern.
Why choose linear trend regression instead of a moving average?
Linear trend regression directly models a sustained rise or fall over time. A moving average smooths recent observations but can lag behind that trend.
Does monthly demand data always require seasonal forecasting?
No. Seasonality requires a recurring pattern, such as December repeatedly running above its deseasonalized level. Monthly observations alone do not establish seasonality.
Practice BAR like the real exam
The free ChatCPA simulator: real exam layout, timed testlets, starting with a question on this topic. No account needed to start.
Open the free simulator →More on Data analytics, regression analysis, and quantitative methods
- Calculate Predicted Cost Using Simple Linear Regression
- Adjusted R-Squared When Adding Regression Predictors
- Correlation vs. Causation: Confounding and Selection Bias
- Sensitivity Analysis vs Scenario Analysis: One Input or Many
All Data analytics, regression analysis, and quantitative methods practice →
Questions from the ChatCPA bank of 17,658 CPA exam questions, each with a written reason for every wrong answer. ChatCPA is built by Nicholas Miller, CPA (Oregon #14907). How these pages are made. Spot an error? Tell us.