We use first differenced logged quarterly series for the GDP of 29 countries and the euro area to assess the need to use non-linear models to describe business cycle dynamic behaviour. Our approach ...is model (estimation)-free, based on testing only. We aim to maximize power to detect non-linearities while, simultaneously, avoiding the pitfalls of data mining. The evidence we find does not support some descriptions because the presence of significant non-linearities is observed for two-thirds of the countries only. Linear models cannot be simply dismissed as they are frequently useful. Contrarily to common knowledge, non-linear business cycle variation does not seem to be a universal, undisputable and clearly dominant stylized fact. This finding is particularly surprising for the U.S. case. Some support for non-linear dynamics for some further countries is obtained indirectly, through unit root tests, but this can hardly be invoked to support non-linearity in classical business cycles.
This paper examines the stationarity of carbon dioxide (CO2) emissions per capita for a set of 36 countries covering the period 1870–2006. We employ recently developed unit root and stationarity ...tests that allow for the mean reverting process to be nonlinear and take into account cross sectional dependence. By grouping countries according to their geographical proximity, the importance of cross sectional dependence in panel unit root and stationarity tests is revealed. Using a recently developed nonlinear panel unit root test, we find strong evidence that the per capita carbon dioxide emissions over the last one hundred and fifty years are stationary. Our nonlinear specification captures the dynamics of the emissions time series data more effectively and we obtain evidence supporting stationarity for all country groups under study.
•The stationarity of the CO2 per capita is examined.•Allow for the mean reverting mechanism to be nonlinear•Linear and nonlinear unit root and stationarity tests are employed.•Non-linearity and cross section dependence are taken into account.
Abstract This paper investigates the degree of persistence of monthly coffee prices of five different groups (ICO Composite, Brazilian Naturals, Other Milds, Colombian Milds, and Robusta) between the ...years 1965 and 2020 by using long memory and fractional integration processes. The results show high degrees of persistence for all coffee groups, rejecting the hypothesis of I(0) stationarity, and showing a lack of mean reversion towards the long-run price level for most coffee groups. The results imply that in the case of an exogenous shock in coffee price levels, strong policy measures must be introduced to regain their original levels.
This study examines the stochastic conditional convergence of sulphur dioxide (SO
2
) emissions using the Residual Augmented Least Squares-Lagrange Multiplier (RALS-LM) unit root test with structural ...breaks. This procedure allows the data to account for trend breaks and nonnormal errors that have been ignored or deemphasized in previous studies. The study finds that per capita sulphur dioxide emissions exhibits stochastic conditional convergence across US states.
Crude oil price behaviour depends on all the events that have the potential to disrupt the flow of oil. We understand that these causes could be geopolitical issues and/or military conflicts in/with ...the producer countries and a problem relating to demand and supply. In the paper we first investigate the statistical properties of the real oil prices as well as its log-transformation, along with the absolute and squared returns values. Then, we also address the following issue: Does the crude oil price behave in the same way before and after a military conflict or geopolitical problem in the producer countries? To answer this question we analyse the real oil prices of West Texas Intermediate (WTI) before and after the different military conflicts and political events that occurred after World War II. For this purpose we use techniques based on unit roots and fractional integration. The empirical results provide evidence of persistence and breaks in the oil prices series and stationary long memory in the absolute returns. However, we do not observe significant differences before and after the conflict and geopolitical events.
•This paper investigates the statistical properties of oil prices.•The paper analyses oil prices before and after a military conflict.•We use techniques based on unit roots and fractional integration.•We detect persistence and breaks in the oil prices series.•We find stationary long memory in the absolute returns.
This study extends the literature on the permanent or transitory nature of shocks to per capita tourist arrivals along several dimensions. First, the study evaluates the nature of shocks to per ...capita tourist arrivals for a global panel of 129 countries. Second, unlike previous studies, we jointly estimate structural changes that represent either abrupt breaks or as a Fourier approximation of smooth breaks along with introducing a factor structure to test for the presence of unit roots in per capita tourist arrivals. Third, contrary to previous studies, our results show that rejection of the null hypothesis of a unit root in per capita tourist arrivals is quite limited compared to other unit root tests that fail to account for cross-correlations. Policy implications of the findings are also discussed.
Commodities prices are pivotal to the mineral investment decision and have a considerable impact on mining companies' financial performance and countries that depend on mineral resources. Therefore, ...understanding the future mineral price movement is critical for revenue-based planning both for the company and the country. In this article, the Autoregressive Distribution Lag (ARDL) model was used to forecast annual gold prices using gold demand, treasury bills rates, and lagged gold prices as covariates. Augmented Dickey Fuller and the Phillips Perron methods were used to test for unit roots and found that all the variables were integrated of order one. Subsequently, the cointegration test was undertaken, which indicated that there is no cointegration between the variables. This entailed application of the short-run version of the ARDL to forecasts and consequent analysis. A Granger causality analysis show that gold demand Granger causes gold price; and that treasury bill rates do not Granger cause gold price. Lastly, the ARDL (4, 4, 2) model, which provides best ARDL forecast results, was evaluated against two other forecasting methods namely stochastic mean reverting, and Autoregressive Integrate Moving Average (ARIMA). Results showed that the ARDL model emerged as a best of all the three forecasting methods to forecast annual gold prices.
•ARDL-ECM was used to forecast gold prices using demand, treasury bill rates and lagged gold prices.•Stationarity test, estimation of model equation, cointegration test, diagnostic tests, short and long run models were used.•ARDL-ECM model results subsequently compared with stochastic Gaussian model and Autoregressive Integrate Moving Average.•The ARDL model emerged as a best of all the three methods to forecast annual gold prices based on error metrics.
Functional principal component analysis (FPCA) has played an important role in the development of functional time series analysis. This note investigates how FPCA can be used to analyze cointegrated ...functional time series and proposes a modification of FPCA as a novel statistical tool. Our modified FPCA not only provides an asymptotically more efficient estimator of the cointegrating vectors, but also leads to novel FPCA‐based tests for examining essential properties of cointegrated functional time series.
We analyze the asymptotic properties of nonparametric cross spectral density estimators when the time series process has a degenerate spectrum at the origin in the frequency domain. Degeneracy at the ...zero frequency caused by over-differencing the series invalidates inferences based on cross spectral methods. We derive the asymptotic mean squared errors of kernel-based cross spectral density estimators at the zero frequency. It is found that degeneracy entails different asymptotics including convergence rates for nonparametric cross spectral density estimators from existing results in time series econometrics literature.