E37 - Prices, Business Fluctuations, and Cycles: Forecasting and Simulation: Models and ApplicationsReturn

Random forest models have recently gained popularity for economic forecasting. Earlier studies demonstrated their potential to provide early warnings of recession and serve as a competitive method to older prediction models. This study offers the first evaluation of the random forest forecast for the Czech economy. The one-step-ahead forecasting results show high accuracy on the Czech data and are proven to outperform forecasts from the Czech Ministry of Finance and the Czech National Bank. The following multi-step random forest forecast, estimated for the next four quarters, shows results similar to those from the central institutions. The main difference stems from the household and industrial confidence variables, which significantly impact on the random forest forecast. The variable-importance analysis further emphasizes the soft variables as valuable determinants for Czech forecasting. Overall, the findings motivate other forecasters to exercise this method.

This paper presents a theoretical model that is suitable for the analysis of price-setting hetero-geneity in a small open economy. The model is based on Benigno and López-Salido's (2006) work and is one of many examples of a dynamic stochastic general equilibrium (DSGE) model that utilizes heterogeneous price stickiness (e.g., Aoki, 2001; Bodenstein et al., 2008). The model allows analyses of a small open economy by extending the existing multisector models using the mechanisms described by Galí and Monacelli (2005). As a result, the model enables monetary policy analyses that take into account existing sectoral differences in the price-setting mechanisms found in an open economy. In the empirical part of the paper, the model is estimated on data for Poland using Bayesian techniques. The results show that the period 1999-2017 saw significant differences in price stickiness and inflation persistence in the sectors that produce food and energy compared with sectors that produce other goods and services.

Global economic policy uncertainty (GEPU) is one of important phenomena in the global economy; it can impact on the overall economic performance and stock market per-formance, regardless of the status of the world economy. Thus, this paper empirically investigates the impact of global economic policy uncertainty on the Malaysian stock market over the period from 10:2003 to 2017:03. Using the GARCH model, the study demonstrates that global policy uncertainty affects the Malaysian stock market negatively. Similarly, the SVAR model also shows results consistent with the GARCH estimation. Nevertheless, the Markov switching estimation uncovers that global policy uncertainty has negative impacts on stock market performance in both low and high volatile market states. The impact is, however, greater during the high volatile state. Hence, the relationship between global economic policy uncertainty and stock market returns tends to be asymmetric. The overall empirical results infer that global economic policy uncertainty has some implications for asset pricing.

This paper evaluates the accuracy of real GDP growth forecasts published in the period 1995-2013 by two Czech institutions: the Ministry of Finance (MF) and the Czech National Bank (CNB). A two-stepped approach is adopted: first a battery of forecasting errors (MAE, RMSE, MASE) is calculated, complementary to evaluation papers already available. Then statistical analysis is carried out by comparing both MF and CNB forecasts with OECD, European Commission and consensus benchmarks (Kruskal-Wallis test), assessing the presence of systemic bias (Wilcoxon test) and determining their incremental improvement (Page trend test). The results show that although some error patterns might suggest performance deficiencies (i.e. during the recession periods), the accuracy of forecasts prepared by both the MF and CNB does not differ significantly from the benchmark forecasts; MF and CNB predictions do not contain systemic bias and their accuracy improves as the horizon shortens. The paper also highlights several methodological shortcomings in the internal evaluations conducted by both institutions, indicating a potential for further improvement.

We compare three filters commonly used for business cycle analysis: the Baxter-King, the Christiano-Fitzgerald and the Hamming window filter. Empirical contribution of the paper is numerical evaluation of the approximation of the ideal band-pass filters in the discussion of the filters' theoretical properties (gain and attenuation within the business cycle frequencies, as well as the leakage in the remaining frequencies). We consider the truncation factor for the BaxterKing filter and the sample size for the latter two. We show that the leakage and attenuation of the Christiano-Fitzgerald and the Hamming window filter perform similarly across the range of chosen sample sizes and better than the Baxter-King filter. Moreover, we apply the filters to data of selected EU countries and point out differences in their estimation of growth business cycles. Our findings indicate that Christiano-Fitzgerald filter and the Hamming window both are appropriate for the identification of a business cycle. The Hamming window filter introduces smaller attenuation near the edges but in case of small samples its approximation of ideal filter is very rough.

In this article we analyse a neoclassical model of inflation. Our aim is to reconstruct the neoclassical theory of inflation to obtain a model which generates non-periodical oscillations of price level. This model is considered to be a realistic approximation of actual price level evolution. We start our analysis with the Fisherian equation of exchange. The assumption on non-variability of the velocity of money circulation parameter is relaxed in favour of dependence on expected inflation. The resulting model of inflation is a two-equation model where price evolution depends on production dynamics which is assumed to be an exogenous variable. After that, the two-equation model is re-formulated as an autonomous system to a model where production dynamics is determined by a Kaldorian type's model. By adding Kaldor's model to the two equation system, we create a four equation model. Both our models are able to generate more complex dynamics, i.e. non-linear cycles and chaos, which we examine by generating time series from numerical example and analyse them with the help of an advanced non-linear method.

This paper examines the exchange rate pass-through to import, producer and consumer prices in South Africa using monthly data covering the period 2000M1 to 2009M5. The study uses innovation accounting tools (impulse response and variance decomposition) within the framework of an unrestricted VAR to examine the degree of pass-through as well as the relative importance of a number of variables in explaining changes in domestic prices. The key findings suggest that after 1 per cent shock to nominal effective exchange rate, the level of CPI increases by 0.125 per cent, giving a pass-through elasticity of 13 per cent. However, the pass-through elasticity of producer price is 20 per cent after 24 months suggesting that favourable shocks to producer price inflation can have considerable moderating effect on CPI inflation.

Financial sector is prone to cyclical movements and procyclicality of the financial system may endanger financial stability, which depends on asset prices and loan losses due to the fact that the deterioration of bank assets through non-performing loans is characteristics of banking distress. This was the case during Japan's lost decade and the Nordic banking crises. Even the classic banking panics of the Great Depression are being revised in the light of new evidence on the fundamental deterioration of bank assets. Much empirical evidence supports the view that balance sheet variables, such as net worth affect investment and produce business cycle dynamics. In an upswing, the greater availability of credit leads to higher asset prices, which then serve as collateral for more borrowing. Relatively unstable development of share prices on the capital market increases equity risk. This paper is based on the presumption that the stability of macro economic environment, less pronounced cyclical movements and insignificant procyclicality between GDP and equity (used as collaterals for credit insurance) lower equity risk. There was proved no significant procyclicality between collaterals and GDP according to low stock market capitalization. And due to the relation that equity risk (as a part of market risk) is determined by unstable development of shares prices, I accepted the hypothesis of low equity risk in the analysed transition economies on the basis of tested procyclicality.

Leading economic indicators have a long tradition in forecasting future economic activity. Recent developments, however, suggest that there is scope for adding extensions to the methodology of forecasting major economic fluctuations. In this paper, the author tries to develop a new model, which would outperform the forecast accuracy of classical leading indicators model. The use of artificial neural networks is proposed here. For demonstration a case study for Slovene economy is included. The main finding is that, at the twelve months forecasting horizon, a stable and improved forecast accuracy could be achieved for in- and out-of-sample data.