<abstract xmlns="http://www.w3.org/1999/xhtml"><p>Croatian Industrial Confidence Indicator (ICI) is one of the measures of mangers’ sentiment about the economic situation in the Croatian manufacturing industry. Since 2005, the ICI has been calculated in accordance with the harmonised European Commission methodology as a simple average of three variables: order books, stocks of finished products and production expectation. It was empirically confirmed that the ICI could predict the direction of change in industrial production more than one month ahead. With the aim of raising the ICI forecasting power, this paper proposes a novel ICI with a different weighting scheme. The empirical analysis is based on monthly data for three standard ICI subcomponents and industrial production expressed as year-on-year growth rates. The data set covers the period from May 2008 to February 2019. Data sources were the European Commission and Eurostat. The newly defined ICI was constructed by using the nonlinear optimisation approach. The weights were determined by minimizing the root mean square error (RMSE) in a simple regression model and by maximizing the correlation coefficient between the ICI and industrial production for various time lags. The results reveal that the newly defined ICI performs better in adapting and following the industrial production growth rate as well as that the dominant component in the ICI is the production expectation.</p></abstract>

Croatian Industrial Confidence Indicator (ICI) is one of the measures of mangers’ sentiment about the economic situation in the Croatian manufacturing industry. Since 2005, the ICI has been calculated in accordance with the harmonised European Commission methodology as a simple average of three variables: order books, stocks of finished products and production expectation. It was empirically confirmed that the ICI could predict the direction of change in industrial production more than one month ahead. With the aim of raising the ICI forecasting power, this paper proposes a novel ICI with a different weighting scheme. The empirical analysis is based on monthly data for three standard ICI subcomponents and industrial production expressed as year-on-year growth rates. The data set covers the period from May 2008 to February 2019. Data sources were the European Commission and Eurostat. The newly defined ICI was constructed by using the nonlinear optimisation approach. The weights were determined by minimizing the root mean square error (RMSE) in a simple regression model and by maximizing the correlation coefficient between the ICI and industrial production for various time lags. The results reveal that the newly defined ICI performs better in adapting and following the industrial production growth rate as well as that the dominant component in the ICI is the production expectation.

Nonlinear optimisation approach to proposing novel Croatian Industrial Confidence Indicator

Croatian Review of Economic, Business and Social Statistics

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Croatian Industrial Confidence Indicator (ICI) is one of the measures of mangers’ sentiment about the economic situation in the Croatian manufacturing...

Nonlinear optimisation approach to proposing novel Croatian Industrial Confidence Indicator

Published Online: Dec 21, 2019

Page range: 17 - 26

Received: Jul 01, 2019

Accepted: Sep 16, 2019

DOI: https://doi.org/10.2478/crebss-2019-0008

Keywords
business and consumer survey, industrial confidence indicator, nonlinear optimisation

© 2019 Mirjana Čižmešija et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Nonlinear optimisation approach to proposing novel Croatian Industrial Confidence Indicator

Published Online: Dec 21, 2019

Page range: 17 - 26

Received: Jul 01, 2019

Accepted: Sep 16, 2019

DOI: https://doi.org/10.2478/crebss-2019-0008

Keywordsbusiness and consumer survey, industrial confidence indicator, nonlinear optimisation

© 2019 Mirjana Čižmešija et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Keywords
business and consumer survey, industrial confidence indicator, nonlinear optimisation