Mid-Term Exam: Sales Forecasting MCQs
Uploaded by
RuchiMid-Term Exam: Sales Forecasting MCQs
Uploaded by
RuchiCommon questions
Powered by AIChoosing different statistical measures affects cross-series forecast comparison due to differences in sensitivity to error scales and units. RMSE provides absolute forecast error magnitudes, emphasizing larger errors, while MAPE standardizes errors into percentage terms, facilitating comparison across differently scaled series. Decision-making is influenced by priority on error sensitivity—RMSE suits when larger errors are critically impactful, while MAPE aids in relative performance assessment across varied units.
RMSE treats large and small forecast errors unequally because it is based on a quadratic error cost function that squares the errors, thus disproportionately penalizing larger errors. This characteristic makes it sensitive to large deviations in forecasts, affecting evaluation by emphasizing the need for accuracy in those predictions with larger mistakes.
The decision between a three-month and a six-month moving average depends on the underlying data cycle and forecasting goals. A three-month average is preferred when data shows short-term patterns or cycles, offering more sensitivity to recent changes, whereas a six-month average provides smoother results for longer-term stability. Preference is determined by data variance and forecast purpose.
Selecting low α values in exponential smoothing for highly random data allows the model to dampen the influence of erratic fluctuations in the data. Lower α values give greater weight to older data points, resulting in smoother forecasts and reducing the impact of random noise, which improves the forecast reliability for volatile series.
Using moving averages for deseasonalizing a time series is effective when the number of periods matches the seasonal cycle, helping smooth out seasonality by averaging over cyclic influences. Careful selection of the period length is critical to accurately reflect seasonal patterns, such as using 4-period averages for quarterly data or 12 for monthly, to ensure the average effectively captures typical seasonal fluctuations.
Holt's smoothing method extends simple exponential smoothing by including an additional equation to account for trends, making it suitable for data with both level and trend components but lacking seasonality. It adjusts forecasts to anticipate future movements by considering both current level changes and trend projections, thus improving adaptability to nonstationary data.
Time-series decomposition involves breaking down a series into its fundamental components: trend, seasonality, and irregularity. This process helps isolate systematic patterns such as trends or cycles, allowing forecasters to focus on individual components for targeted predictions. It enhances forecast accuracy by enabling adjustments that account separately for trend shifts, seasonal fluctuations, and random noise, facilitating more precise modeling.
Seasonal effects impact forecast accuracy when data exhibits regular cyclical patterns. For example, using previous month's sales data to forecast January may lead to inaccuracies if data is seasonal. Mitigating these effects involves adjusting forecasts for seasonality using methods like seasonal decomposition or incorporating seasonal indices to normalize data based on expected fluctuations.
For forecasting sales of new products, methods like the Delphi Method and customer surveys are more applicable as they do not require historical sales data and rely on expert judgment and consumer intention, respectively. Time series techniques, which require historical data, are less suitable for new products due to the absence of such data.
The Delphi Method is characterized by gathering anonymous opinions from a panel of experts to generate forecasts. It is effective for subjective forecasting as it avoids the influence of dominant personalities, facilitates unbiased judgement aggregation, and leads to consensus through a series of iterations and feedback.