Which of the following lists the five models of Time Series?

• A. Naive forecasting, simple moving average forecasting, weighted moving average, exponential smoothing, and linear trend
• B. Simple regression and multiple regression
• C. MAD, MAPE, MSE
• D. Qualitative methods

Answer: (A)

Time series forecasting relies on patterns in time-ordered data, using past observations to predict future values. The five models listed are classic time-series forecasting methods: Naive forecasting uses the most recent observation as the forecast for the next period; a simple moving average averages a fixed number of most recent observations to smooth out fluctuations; a weighted moving average also averages past observations but assigns greater weight to more recent data; exponential smoothing applies exponentially decreasing weights to past observations, giving more emphasis to newer data and forming a family of methods that can address level and sometimes trend or seasonality; and a linear trend model fits a straight line to the data over time to forecast future values. These are all inherently time-ordered forecasting methods. The other options don’t fit as a set of time-series models: regression approaches (simple or multiple) rely on predictors and aren’t inherently time-ordered models; error measures (MAD, MAPE, MSE) assess forecast accuracy rather than provide forecasting models; qualitative methods rely on judgment rather than quantitative time-series techniques.