Predicting Inflation by the Main Inflationary Factors: Performance of TVP-VAR and VAR-NN models

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F62156489%3A43110%2F15%3A43906588" target="_blank" >RIV/62156489:43110/15:43906588 - isvavai.cz</a>

Výsledek na webu

<a href="http://mme2015.zcu.cz/downloads/MME_2015_proceedings.pdf" target="_blank" >http://mme2015.zcu.cz/downloads/MME_2015_proceedings.pdf</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Predicting Inflation by the Main Inflationary Factors: Performance of TVP-VAR and VAR-NN models

Popis výsledku v původním jazyce

A suitable way for forecasting inflation is to do it using main inflationary factors. Such factors can be sorted to domestic and foreign sets. One-way and two- way relations between them and inflation can be considered. Therefore, vector autoregressive model (VAR) seems to be a proper tool for modelling the reality. However, basic VAR model can suffer from insufficient forecasting performance caused by its linear nature. We employ two nonlinear vector autoregressive alternatives for predicting inflation: Time-Varying Parameter VAR model with stochastic volatility and VAR Neural Network model. In both cases we select the specification with the best combination of inflationary factors. Neural Networks are flexible tool which can be easily adjusted to anautoregressive form. Resulting VAR-NN models produce accurate inflation forecasting, but they take essential information mainly from the previous inflation observations and ignore the other series. Compared to that, TVP-VAR model is a sta

Název v anglickém jazyce

Predicting Inflation by the Main Inflationary Factors: Performance of TVP-VAR and VAR-NN models

Popis výsledku anglicky

A suitable way for forecasting inflation is to do it using main inflationary factors. Such factors can be sorted to domestic and foreign sets. One-way and two- way relations between them and inflation can be considered. Therefore, vector autoregressive model (VAR) seems to be a proper tool for modelling the reality. However, basic VAR model can suffer from insufficient forecasting performance caused by its linear nature. We employ two nonlinear vector autoregressive alternatives for predicting inflation: Time-Varying Parameter VAR model with stochastic volatility and VAR Neural Network model. In both cases we select the specification with the best combination of inflationary factors. Neural Networks are flexible tool which can be easily adjusted to anautoregressive form. Resulting VAR-NN models produce accurate inflation forecasting, but they take essential information mainly from the previous inflation observations and ignore the other series. Compared to that, TVP-VAR model is a sta

Druh

D - Stať ve sborníku

CEP obor

BB - Aplikovaná statistika, operační výzkum

OECD FORD obor

—

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2015

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Mathematical Methods in Economics 2015: Conference Proceedings

ISBN

978-80-261-0539-8

ISSN

—

e-ISSN

—

Počet stran výsledku

6

Strana od-do

133-138

Název nakladatele

Západočeská univerzita

Místo vydání

Plzeň

Místo konání akce

Cheb

Datum konání akce

9. 9. 2015

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—