Sub: Statistics Topic: Forecasting
Several techniques are available to forecast time-series data that are stationary or that include nosignificant trend, cyclical, or seasonal effects. These techniques are often referred to as
produce forecasts based on “
” the irregular fluctuation
effects in the time-series data. Three general categories of smoothing techniques are presented here:
Naive forecasting models
are simple models in which it is assumed that the more recent timeperiods of data represent the best predictions or forecasts for future outcomes. Naive modelsdo not take into account data trend, cyclical effects, or seasonality. For this reason, naivemodels seem to work better with data that are reported on a daily or weekly basis or insituations that show no trend or seasonality. The simplest of the naive forecasting methods isthe model in which the forecast for a given time period is the value for the previous timeperiod.
= the forecast value for time period t
= the value for time period t
1Many naive model forecasts are based on the value of one time period. Often such forecastsbecome a function of irregular fluctuations of the data; as a result, the forecasts are
Using averaging models, a forecaster enters information from several timeperiods
into the forecast and “smoothes” the data.
are computed byaveraging data from several time periods and using the average as the forecast for the next