Figure 12.1 lists the eight most common time-series models, segregated by seasonality and trend. For instance, if the data variable has no trend or seasonality, then a single moving average model or a single exponential-smoothing model would suffice. However, if seasonality exists but no discernable trend is present, either a seasonal additive or seasonal multiplicative model would be better, and so forth. The following sections explore these models in more detail. These computational examples use monthly data with a seasonality of 4. However, in practice, any periodicity can be used (e.g., minutes, hours, days, months, quarters, years, untimed, etc.) and any seasonality period can be applied (e.g., 1 for annual data, 12 for monthly data, 4 for quarterly data, 24 for hourly data, etc.).
Figure 12.1: The Eight Most Common Time-Series Methods
