Abstract

Let us try to predict a time series. The goal here is to establish a function based on observed values (time series data). Using

$$x_1,x_2,x_3,...,x_t,~~~~~~~(4.1)$$

we attempt to obtain a function

$$x_t = f (x_{t-1},x_{t-2},x_{t-3},x_{t-4},...,x_{t-M})~~~~~~~ (4.2)$$

that can be used to predict current data x t from previously observed data. The reader should note that the arguments of this function do not include a time variable t. This is to avoid deriving a time series prediction function that is dependent on absolute time.

Keywords

Genetic Programming Minimum Description Length Time Series Prediction Linear Genetic Program Kernel Model 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer Berlin Heidelberg 2012

Authors and Affiliations

  1. 1.School of Engineering, Dept. Information & Communication EngineeringUniversity of TokyoTokyoJapan
  2. 2.School of Frontier Sciences, Department of Frontier InformaticsThe University of TokyoTokyoJapan

Personalised recommendations