1|Page
QMB 3200 Business Statistics Final Exam |
Time Series, Forecasting, Regression,
Hypothesis Testing | Multiple Choice &
Open-Ended Q&A | Verified Answers
Exam Structure:
Subject: Business Statistics – Time Series & Forecasting (QMB 3200)
Source: QMB 3200 Final Exam – Verified Answers
Format: Multiple Choice & Open-Ended Q&A
1. Which of the following is NOT present in a time series?
Correct Answer: Operational variations
Rationale:
1. Time series components typically include trend, seasonal, cyclical, and
irregular (random) variations.
2. Operational variations are not a standard component of time series
decomposition.
3. This term may refer to variations caused by changes in business
operations, which are not inherently time-based.
4. Recognizing standard components is essential for selecting appropriate
forecasting methods.
2. The difference between the actual time series value and the forecast
is called:
Correct Answer: Forecast error
Rationale:
1. Forecast error = Actual – Forecast.
2. Errors can be positive (underforecast) or negative (overforecast).
, 2|Page
3. Forecast errors are used to calculate accuracy metrics (MAE, MSE, MAPE).
4. Minimizing forecast error is the goal of forecasting methods.
3. What type of analysis aims to discover a pattern in historical data or
a time series and then extrapolate that pattern into the future?
Correct Answer: Time series analysis
Rationale:
1. Time series analysis uses only past values of the variable to predict future
values.
2. Assumes that historical patterns (trend, seasonality, cycles) will continue.
3. Contrasts with causal forecasting (uses external predictors).
4. Methods include moving averages, exponential smoothing, and
decomposition.
4. The average of the absolute values of the forecast errors is called:
Correct Answer: Mean absolute error (MAE)
Rationale:
1. MAE = (1/n) Σ|Actual – Forecast|.
2. MAE is in the same units as the original data (easy to interpret).
3. Less sensitive to outliers than MSE.
4. Lower MAE indicates more accurate forecasts.
5. What is the component of a time series model that is attributable to
multi-year cycles in the time series?
Correct Answer: The cyclical component
Rationale:
1. Cyclical patterns last longer than one year (e.g., 5-10 year business
cycles).
2. Distinguished from seasonal patterns (within one year).
3. Often tied to economic expansions and contractions.
4. Cyclical components are less regular and harder to predict than seasonal
patterns.
6. Time series regression refers to the use of regression analysis when
the independent variable is:
Correct Answer: Time
, 3|Page
Rationale:
1. In time series regression, time (t) is the predictor variable.
2. Model: yₜ = b₀ + b₁t + ε (linear trend).
3. Can also include polynomial terms (t²) for curvature.
4. Used to model and forecast trend patterns.
7. The method that uses the average of the most recent k data values
in the time series as the forecast for the next period is called:
Correct Answer: Moving averages
Rationale:
1. Moving average forecast = (Yₜ + Yₜ₋₁ + … + Yₜ₋ₖ₊₁) / k.
2. Smooths out short-term fluctuations.
3. Larger k produces smoother forecasts but less responsiveness.
4. Best for stationary or nearly stationary series.
8. If the historical data are restricted to past values of the variable to
be forecast, the forecasting procedure is called a:
Correct Answer: Time series method
Rationale:
1. Time series methods use only the history of the variable being forecast.
2. No external predictors are used.
3. Examples: moving averages, exponential smoothing, decomposition.
4. Contrast with causal methods (use external variables).
9. What forecasting method uses a weighted average of past time
series values as the forecast; it is a special case of weighted moving
averages in which we select only one weight – the weight for the most
recent observation?
Correct Answer: Exponential smoothing
Rationale:
1. Exponential smoothing: Fₜ₊₁ = αYₜ + (1-α)Fₜ.
2. All past observations are included with exponentially decreasing weights.
3. Only one parameter (α) needs to be selected.
4. Special case of weighted moving averages with geometrically declining
weights.
QMB 3200 Business Statistics Final Exam |
Time Series, Forecasting, Regression,
Hypothesis Testing | Multiple Choice &
Open-Ended Q&A | Verified Answers
Exam Structure:
Subject: Business Statistics – Time Series & Forecasting (QMB 3200)
Source: QMB 3200 Final Exam – Verified Answers
Format: Multiple Choice & Open-Ended Q&A
1. Which of the following is NOT present in a time series?
Correct Answer: Operational variations
Rationale:
1. Time series components typically include trend, seasonal, cyclical, and
irregular (random) variations.
2. Operational variations are not a standard component of time series
decomposition.
3. This term may refer to variations caused by changes in business
operations, which are not inherently time-based.
4. Recognizing standard components is essential for selecting appropriate
forecasting methods.
2. The difference between the actual time series value and the forecast
is called:
Correct Answer: Forecast error
Rationale:
1. Forecast error = Actual – Forecast.
2. Errors can be positive (underforecast) or negative (overforecast).
, 2|Page
3. Forecast errors are used to calculate accuracy metrics (MAE, MSE, MAPE).
4. Minimizing forecast error is the goal of forecasting methods.
3. What type of analysis aims to discover a pattern in historical data or
a time series and then extrapolate that pattern into the future?
Correct Answer: Time series analysis
Rationale:
1. Time series analysis uses only past values of the variable to predict future
values.
2. Assumes that historical patterns (trend, seasonality, cycles) will continue.
3. Contrasts with causal forecasting (uses external predictors).
4. Methods include moving averages, exponential smoothing, and
decomposition.
4. The average of the absolute values of the forecast errors is called:
Correct Answer: Mean absolute error (MAE)
Rationale:
1. MAE = (1/n) Σ|Actual – Forecast|.
2. MAE is in the same units as the original data (easy to interpret).
3. Less sensitive to outliers than MSE.
4. Lower MAE indicates more accurate forecasts.
5. What is the component of a time series model that is attributable to
multi-year cycles in the time series?
Correct Answer: The cyclical component
Rationale:
1. Cyclical patterns last longer than one year (e.g., 5-10 year business
cycles).
2. Distinguished from seasonal patterns (within one year).
3. Often tied to economic expansions and contractions.
4. Cyclical components are less regular and harder to predict than seasonal
patterns.
6. Time series regression refers to the use of regression analysis when
the independent variable is:
Correct Answer: Time
, 3|Page
Rationale:
1. In time series regression, time (t) is the predictor variable.
2. Model: yₜ = b₀ + b₁t + ε (linear trend).
3. Can also include polynomial terms (t²) for curvature.
4. Used to model and forecast trend patterns.
7. The method that uses the average of the most recent k data values
in the time series as the forecast for the next period is called:
Correct Answer: Moving averages
Rationale:
1. Moving average forecast = (Yₜ + Yₜ₋₁ + … + Yₜ₋ₖ₊₁) / k.
2. Smooths out short-term fluctuations.
3. Larger k produces smoother forecasts but less responsiveness.
4. Best for stationary or nearly stationary series.
8. If the historical data are restricted to past values of the variable to
be forecast, the forecasting procedure is called a:
Correct Answer: Time series method
Rationale:
1. Time series methods use only the history of the variable being forecast.
2. No external predictors are used.
3. Examples: moving averages, exponential smoothing, decomposition.
4. Contrast with causal methods (use external variables).
9. What forecasting method uses a weighted average of past time
series values as the forecast; it is a special case of weighted moving
averages in which we select only one weight – the weight for the most
recent observation?
Correct Answer: Exponential smoothing
Rationale:
1. Exponential smoothing: Fₜ₊₁ = αYₜ + (1-α)Fₜ.
2. All past observations are included with exponentially decreasing weights.
3. Only one parameter (α) needs to be selected.
4. Special case of weighted moving averages with geometrically declining
weights.
