The apparatus for performing short-term prediction of the disordered time series data reconstructs the attractor of the time series data of the target data in a multi-dimensional state space and a variable determinant which changes each variable value according to the dynamics of the observed disordered time series data, It consists of a predictor which predicts a near future target data value by using a vector neighboring the data vector containing the target data. Thus, it is possible to quickly and accurately perform short-term prediction of disordered time series data and to cope with dynamic changes in time series data.
展开▼