Detail publikace

Identification of time-varying model using wavelet approach and AR process

Originální název

Identification of time-varying model using wavelet approach and AR process

Anglický název

Identification of time-varying model using wavelet approach and AR process

Jazyk

en

Originální abstrakt

The paper aim is to give recommendation for work with time frequency modeling of macroeconomic time series on the basis of comparative study. We investigate wavelet analysis and time-varying autoregressive process. We follow two main areas of focus - sample size for available data set and its shortening and optimization of parameters of mentioned methods. In case of time-varying autoregressive process we investigate optimization of parameters such as lag length, windowing and overlap. In wavelet analysis approach we investigate the type of wave and scale. Performance of methods is presented on the gross domestic product data of USA, United Kingdom and Korea. These representatives were chosen from the perspective of available sample size and from the reason the country represent economy in different geographic area. An advantage of wavelet analysis is better time resolution. An autoregressive process provides better frequency resolution, but it is quite sensitive to sample size.

Anglický abstrakt

The paper aim is to give recommendation for work with time frequency modeling of macroeconomic time series on the basis of comparative study. We investigate wavelet analysis and time-varying autoregressive process. We follow two main areas of focus - sample size for available data set and its shortening and optimization of parameters of mentioned methods. In case of time-varying autoregressive process we investigate optimization of parameters such as lag length, windowing and overlap. In wavelet analysis approach we investigate the type of wave and scale. Performance of methods is presented on the gross domestic product data of USA, United Kingdom and Korea. These representatives were chosen from the perspective of available sample size and from the reason the country represent economy in different geographic area. An advantage of wavelet analysis is better time resolution. An autoregressive process provides better frequency resolution, but it is quite sensitive to sample size.

BibTex


@inproceedings{BUT116326,
  author="Eva {Klejmová} and Jitka {Poměnková}",
  title="Identification of time-varying model using wavelet approach and AR process",
  annote="The paper aim is to give recommendation for work with time frequency modeling of macroeconomic time series on the basis of comparative study. We investigate wavelet analysis and time-varying autoregressive process. We follow two main areas of focus - sample size for available data set and its shortening and optimization of parameters of mentioned methods. In case of time-varying autoregressive process we investigate optimization of parameters such as lag length, windowing and overlap. In wavelet analysis approach we investigate the type of wave and scale. Performance of methods is presented on the gross domestic product data of USA, United Kingdom and Korea. These representatives were chosen from the perspective of available sample size and from the reason the country represent economy in different geographic area. An advantage of wavelet analysis is better time resolution. An autoregressive process provides better frequency resolution, but it is quite sensitive to sample size.",
  address="University of West Bohemia, Plzeň",
  booktitle="33rd International Conference Mathematical Methods in Economics MME 2015 - Conference Proceedings",
  chapter="116326",
  howpublished="online",
  institution="University of West Bohemia, Plzeň",
  year="2015",
  month="september",
  pages="665--670",
  publisher="University of West Bohemia, Plzeň",
  type="conference paper"
}