Time Series Analysis Excel Template

Description
Time series analysis is useful for forecasting based on the patterns underlying the past data. There are four main components:

1) Trend: A long-term movement regarding time series that can be upward, downward, or stationary (e.g. upward trend in population growth)

2) Cyclical: A pattern typically observed over two or more years caused by circumstances that repeat in cycles (e.g. economic cycles)

3) Seasonal: Variations within a year that generally depend on the weather, habits of consumers etc.

5) Irregular components: Events with unpredictable influences on the time series.


There are two main types of models depending on how the previous four components are included:

1) Y(t)=T(t) x S(t) x C(t) x I(t)
Multiplicative models: the four components are multiplied, and in this case we assume that the components can affect each other.

2) Y(t)=T(t) + S(t) + C(t) + I(t)
Additive models: we make the assumption that the components are independent.

Stationarity is another important element of time series. A stationary process is when an event is influenced by a previous event or events. E.g. a high temperature today gives the impression tomorrow the temperature is likely to be high too.

Many models for time series analysis exist, yet the most commonly-used is the ARIMA (autoregressive integrated moving average). There are several variations of this as well as non-linear models. Nevertheless, linear models such as the ARIMA are widely used due to their simplicity in both execution and understanding.

A good time series connotes several exploratory analyses and model validation that required statistical knowledge and experience. This template includes a simplification of a time series model in which seasonality and trends are isolated to forecast future sales. Data can be gathered at every instance of time (continuous time series), or at discrete points of time (discrete time series).