Time Series forecasting is based on which assumption?

Time Series forecasting is based on which assumption?

• A. The future is unrelated to the past
• B. The past does not influence the forecast
• C. The future is an extension of the past
• D. The future is the average of past forecasts

Answer: (C)

Time series forecasting rests on the idea that the future is an extension of the past. In other words, patterns we see in historical data—such as trends, seasonality, and the way observations relate to recent values—tend to continue into the future. This allows models to learn from past observations and project those patterns forward, assuming the underlying data-generating process remains similar, at least for a while. If the future were unrelated to what happened before, there would be nothing for the model to learn from and forecasts would have no basis. Likewise, if the past didn’t influence the forecast, information from historical data would be useless. The notion isn’t that forecasts come from averaging previous forecasts; instead, they come from a model that captures the structure in the data so past behavior can inform future values. For example, recurring seasonal sales or a persistent upward trend informs how we predict next quarter’s numbers.