Time and frequency analysis of daily-based nexus between global CO2 emissions and electricity generation nexus by novel WLMC approach.

The studies have focused on changes in CO2 emissions over different periods, including the COVID-19 pandemic. Even if CO2 emissions are temporarily reduced during the pandemic according to annual figures, this may be misleading. Considering annual figures is important to understand the overall trend, but using data with much higher frequency (e.g., daily) is much better suited to investigate dynamic relationships and external effects. Therefore, this study comprehensively analyzes the association between CO2 emissions and disaggregated electricity generation (EG) sources across the globe by employing the novel wavelet local multiple correlation (WLMC) approach on daily data from 1st January 2020 to 31st March 2023. The results demonstrate that (1) based on the main statistics, daily CO2 emissions range between 69 MtCO2 and 116 MtCO2, indicating that there is an oscillation, but no sharp changes over the analyzed period. (2) based on the baseline regression using the dynamic ordinary least squares (DOLS) approach, the constructed estimation models have a high predictive ability of CO2 emissions, reaching ~ 94%; (3) in the further analysis employing the WLMC approach, there are significant externalities between EG resources, which affect CO2 emissions. The results present novel insights about time- and frequency-varying effects as well as a disaggregated analysis of the effect of EG on CO2 emissions, demonstrating the significance of the energy transition towards clean sources around the world.

Dynamic link between central bank reserves, credit default swap spreads, and foreign exchange rates: Evidence from Turkey by time series econometrics.

In this study, dynamic links between central bank reserves (CBR), credit default swap (CDS) spreads, and foreign exchange (FX) rates are investigated. So, Turkey, which is a negative outlier country among other peer emerging countries, is examined by considering recent developments on these indicators. In doing so, the study covers relatively high frequency (i.e., weekly) data from January 2, 2004 to November 12, 2021, performs various econometric approaches as Wavelet Coherence (WC), Quantile-on-Quantile Regression (QQR), and Granger Causality in Quantiles (GCQ) as main models, and applies Toda-Yamamoto (TY) causality and Quantile Regression (QR) for the robustness. The results show that (i) there is a time-frequency dependency between the CBR, CDS spreads, and FX rates; (ii) a bidirectional link exists between the CBR and FX rates; between the FX rates and CDS spreads; and between the CDS spreads and CBR; (iii) the link exists in most quantiles except for some lower and middle quantiles for some indicators; (iv) explanatory effect of the indicators on each other varies based on quantiles; (v) the robustness of the results are validated by the TY causality test for the WC model and by the QR approach for the QQR model. The results suggest the significance of the CBR for the FX rates, the FX rates for the CDS spreads, and the CDS spreads for the CBR.

Estimation performance comparison of machine learning approaches and time series econometric models: evidence from the effect of sector-based energy consumption on CO2 emissions in the USA.

By considering the existence of two separate analysis families and the usage of different data frequencies, this study aims to examine the effect of method choice, data frequency, and sector-based energy consumption on carbon dioxide (CO2) emissions by performing machine learning (ML) algorithms and time series econometric (TS) models simultaneously. In this situation, the study examines the United States (USA), considers sector-based energy consumption indicators as explanatory variables, uses monthly and yearly data between January 1973 and December 2021, estimates CO2 emissions, and compares the estimation performance of the models. The empirical findings reveal that (i) the ML algorithms outperform the TS models based on R2 and goodness of fit criteria; (ii) the estimation performance of the models increases with the high-frequency (i.e., monthly) data; (iii) the ML algorithms perform much better in case of high-frequency usage; (iv) some thresholds identify the effects of the sector-based energy consumption indicators on the CO2 emissions; (v) electric power and transportation sectors are the most important sectors in the estimation of the CO2 emissions for monthly and yearly data, respectively. Hence, the study provides to help the understanding role of method choice, data frequency, and sector-based energy consumption for the estimation of CO2 emissions. Based on the results, this study proposes that US policymakers should consider the ML algorithms, use higher-frequency data, and include sector-based energy consumption indicators to have a better estimation of CO2 emissions.

Estimation of Impacts of Global Factors on World Food Prices: A Comparison of Machine Learning Algorithms and Time Series Econometric Models.

It is a well-felt recent phenomenal fact that global food prices have dramatically increased and attracted attention from practitioners and researchers. In line with this attraction, this study uncovers the impact of global factors on predicting food prices in an empirical comparison by using machine learning algorithms and time series econometric models. Covering eight global explanatory variables and monthly data from January 1991 to May 2021, the results show that machine learning algorithms reveal a better performance than time series econometric models while Multi-layer Perceptron is defined as the best machine learning algorithm among alternatives. Furthermore, the one-month lagged global food prices are found to be the most significant factor on the global food prices followed by raw material prices, fertilizer prices, and oil prices, respectively. Thus, the results highlight the effects of fluctuations in the global variables on global food prices. Additionally, policy implications are discussed.