A time series is a series of data points indexed in time order. Most commonly, a time series is a sequence taken at successively equally spaced points in time. Forecasting is about predicting the future (an event of a trend) and as far I am concerned, statistical forecasting implies the use of statistics based on historical data to project what could happen in the future (e.g., sales, GDP, housing prices, etc.).

Perspectives on Forecasting

: Given a specific problem and set of existing methods, how do we solve the problem best? : Given several different forecasting problems, which methods should we have to address them?

Some Common Forecasting Problems

We list some common forecasting problems in the table below and we examine them on the various aspects, including, * Data size and people running the analysis * Forecasting horizon * Time granularity * Cadence of forecast * Human interaction * Other notes