<abstract xmlns="http://www.w3.org/1999/xhtml"><p> As one of the most important indicator for monitoring the production in industry as well as for directing investment decisions, industrial production plays important role within growth perspectives. Not only does the composition and/or fluctuation of the goods produced indicate the course of economic activity but it also reflects the changes in cyclical development of the economy thereby providing opportunity to macro-manage with early signs of (short-term) turning-points and (long-term) trend variations. In this paper, we compare univariate autoregressive integrated moving average (ARIMA) models of the Croatian industrial production and its subsectors in order to evaluate their forecasting features within short and long-term data evolution. The aim of this study is not to forecast industrial production but to analyze the out-of-sample predictive performance of ARIMA models on aggregated and disaggregated level inside different forecasting horizons. Our results suggest that ARIMA models do perform very well over the whole rage of the prediction horizons. It is mainly because univariate models often improve the predictive ability of their single component over the short horizons. In that manner ARIMA modelling could be used at least as a benchmark for more complex forecasting methods in predicting the movements of industrial production in Croatia.</p></abstract>

As one of the most important indicator for monitoring the production in industry as well as for directing investment decisions, industrial production plays important role within growth perspectives. Not only does the composition and/or fluctuation of the goods produced indicate the course of economic activity but it also reflects the changes in cyclical development of the economy thereby providing opportunity to macro-manage with early signs of (short-term) turning-points and (long-term) trend variations. In this paper, we compare univariate autoregressive integrated moving average (ARIMA) models of the Croatian industrial production and its subsectors in order to evaluate their forecasting features within short and long-term data evolution. The aim of this study is not to forecast industrial production but to analyze the out-of-sample predictive performance of ARIMA models on aggregated and disaggregated level inside different forecasting horizons. Our results suggest that ARIMA models do perform very well over the whole rage of the prediction horizons. It is mainly because univariate models often improve the predictive ability of their single component over the short horizons. In that manner ARIMA modelling could be used at least as a benchmark for more complex forecasting methods in predicting the movements of industrial production in Croatia.

Forecasting Capacity of ARIMA Models; A Study on Croatian Industrial Production and its Sub-sectors

Zagreb International Review of Economics and Business

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

As one of the most important indicator for monitoring the production in industry as well as for directing investment decisions, industrial production plays...

Forecasting Capacity of ARIMA Models; A Study on Croatian Industrial Production and its Sub-sectors

Pubblicato online: 09 giu 2017

Pagine: 81 - 99

DOI: https://doi.org/10.1515/zireb-2017-0009

Parole chiave
industrial production, industrial sub-sectors, cycles, ARIMA, forecasting, Croatia

© 2017

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

Forecasting Capacity of ARIMA Models; A Study on Croatian Industrial Production and its Sub-sectors

Pubblicato online: 09 giu 2017

Pagine: 81 - 99

DOI: https://doi.org/10.1515/zireb-2017-0009

Parole chiaveindustrial production, industrial sub-sectors, cycles, ARIMA, forecasting, Croatia

© 2017

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

Parole chiave
industrial production, industrial sub-sectors, cycles, ARIMA, forecasting, Croatia