a review on two interval-valued time series models

Interval-valued data as one important class of symbolic data include observationsdefined by two lower and upper limit points. in this paper, two approachesare presented to forecast the interval-valued time series which are expressed based onthe linear auto-regressive integrated moving average models and a nonlinear artificialneural network model. in each case, two models are fitted on the mid-point and therange of the interval-valued time series in the learning sets. forecasting the lower andupper bounds of the intervals is accomplished by combining the forecasts of the midpointand the range of the intervals. applicability of artificial neural network model asthe alternative forecasting model for interval-valued time series is investigated basedon the mean squared error in the framework of a monte carlo experiment.