# Efficiency of some algorithms for prediction in finite stationary time series

Kybernetika (2004)

- Volume: 40, Issue: 6, page [703]-714
- ISSN: 0023-5954

## Access Full Article

top## Abstract

top## How to cite

topRanocha, Pavel. "Efficiency of some algorithms for prediction in finite stationary time series." Kybernetika 40.6 (2004): [703]-714. <http://eudml.org/doc/33730>.

@article{Ranocha2004,

abstract = {Important characteristics of any algorithm are its complexity and speed in real calculations. From this point of view, we analyze some algorithms for prediction in finite stationary time series. First, we review results developed by P. Bondon [1] and then, we derive the complexities of Levinson and a new algorithm. It is shown that the time needed for real calculations of predictions is proportional to the theoretical complexity of the algorithm. Some practical recommendations for the selection of the best algorithm are given.},

author = {Ranocha, Pavel},

journal = {Kybernetika},

keywords = {stationary time series; multistep prediction; Levinson’s algorithm; innovations algorithm; stationary time series; multistep prediction; Levinson's algorithm; innovations algorithm},

language = {eng},

number = {6},

pages = {[703]-714},

publisher = {Institute of Information Theory and Automation AS CR},

title = {Efficiency of some algorithms for prediction in finite stationary time series},

url = {http://eudml.org/doc/33730},

volume = {40},

year = {2004},

}

TY - JOUR

AU - Ranocha, Pavel

TI - Efficiency of some algorithms for prediction in finite stationary time series

JO - Kybernetika

PY - 2004

PB - Institute of Information Theory and Automation AS CR

VL - 40

IS - 6

SP - [703]

EP - 714

AB - Important characteristics of any algorithm are its complexity and speed in real calculations. From this point of view, we analyze some algorithms for prediction in finite stationary time series. First, we review results developed by P. Bondon [1] and then, we derive the complexities of Levinson and a new algorithm. It is shown that the time needed for real calculations of predictions is proportional to the theoretical complexity of the algorithm. Some practical recommendations for the selection of the best algorithm are given.

LA - eng

KW - stationary time series; multistep prediction; Levinson’s algorithm; innovations algorithm; stationary time series; multistep prediction; Levinson's algorithm; innovations algorithm

UR - http://eudml.org/doc/33730

ER -

## NotesEmbed ?

topTo embed these notes on your page include the following JavaScript code on your page where you want the notes to appear.