Energy Economics

Abstract of associated article: This study adds to the existing literature on oil price–US stock nexus in three ways. First, it employs the VARMA–AGARCH model developed by McAleer et al. (2009) within the context of BEKK framework using West Texas Intermediate (WTI) and Brent as proxies for oil market and S&P stocks as a proxy for US stock market. Secondly, it modifies the model to include endogenously determined structural break using the general structure for analyzing breaks with unit roots in Perron (2006). Third, it uses the adopted model to compute optimal portfolio weight and hedge ratios between oil price and US stocks using different sample data based on the break date. On average, our empirical evidence suggests a significant positive return spillover from US stock market to oil market and bi-directional shock spillovers between the two markets. In addition, there is significant own asymmetric shock effect in both markets while volatility spillover from oil market to stock market became pronounced after the break which coincides with the period of global economic slowdown. Similarly, the results of portfolio management differ across the sample data. More importantly, we find that ignoring structural break when it exists may exaggerate hedging effectiveness.

Abstract of associated article: We analyze production and investment decisions of an agent in industrial activities that are characterized by two forms of uncertainty: demand uncertainty (in terms of number of buyers) and competitive effect uncertainty (in terms of other energy resource). We apply our model on the bioenergy industries. We compare the case of an ambiguity neutral agent with that of an ambiguity averse agent. We show that the investment decision of an agent depends on the effects of both the capital investment and the level of production on the cost and the uncertainty the agent is confronted with. Moreover, we find that ambiguity aversion tends to decrease the agent's optimal levels of production and investment. Our numerical analysis of the French case illustrates the different effects associated with demand uncertainty and competitive effect uncertainty.

Stata synthetic control method programs and required data.

Our dataset consists of monthly data of the oil supply, global demand, domestic demand, real oil price (ROP), import prices, producer prices and consumer prices from January 1999 to December 2016. To reflect the changes in oil supply, world oil production (WOP) provided by the U.S. Energy Information Administration (EIA) is selected, and the data are sourced from the Wind database. Because the Kilian economic index (KI), which was developed by Kilian (2009), can effectively reflect the global economic conditions (Klovland, 2002), following Gupta and Modise (2013) and Hu et al. (2017), we use this index to measure global demand. The year-on-year growth rate of China’s industrial added value (IAV) can effectively reflect China’s economic conditions; thus, following Tan et al. (2015) and Li et al. (2017), the IAV is chosen to represent the domestic demand. Following Gupta and Modise (2013) and Kim et al. (2017), the ROP is defined as the U.S. refiners’ acquisition cost for imported crude oil, for which data are obtained from the U.S. Department of Energy. Based on Shi and Zhao (2016) and Chen and Liao (2017), the IMPI is selected as the proxy for China’s inflation at the import stage. This index measures the changes in China’s imported commodity prices. Following Shi and Zhao (2016) and Zhu et al. (2019), the PPI is chosen to reflect China’s inflation at the production stage. This index measures the changes in the ex-factory price of industrial products. To analyze the effects of oil price shocks on China’s inflation at the consumption stage, we consider the CPI in our analysis. The data of IMPI, PPI and CPI are sourced from the Wind database.

The dataset is composed of the final energy consumption by sector (industry, passenger transport, freight transport, residential, services, and agriculture) and sub-sector/end-use (e.g., chemical industry, cars, truck and light vehicles, space heating, etc.) of Finland. In addition, data regarding passengers and goods traffic, the number of households, the stock of dwellings permanently occupied, the floor area of dwellings, and CO2 emissions are collected. The primary data source is the Odyssee database. The Odyssee data are complemented with data regarding the value added and energy dependence of Finland taken from the World Bank and Eurostat database, respectively. The data of Finland cover the period from 2005 to 2015.

Emissions and Macroeconomic Data underlying research on the paper "Emissions and Economic Development in Commodity Exporting Countries"

The zip-file includes (1) EViews code for replicating the results in the paper; (2) EViews workfile with all of the results and some data series that are so far freely available; and (3) a Readme file with additional information about the files and data series.

Abstract of associated article: Using data on per capita CO2 emissions and per capita real GDP from fifteen countries, spanning the period 1960–2013, this paper tests the validity of the Environmental Kuznets Curve (EKC) using both panel-based and time-series-based methodological approaches of cointegration. Given that the EKC hypothesis postulates an inverted U-shaped relationship between emissions and output, the study tests for cointegration between per capita CO2 emissions, per capita real GDP and the squared values of per capita real GDP. The evidence from panel cointegration methodologies is mixed. This result might arise due to time dependence of cointegrating coefficients. The time-varying cointegration approaches provide strong evidence in favor of time-varying cointegration parameters. Furthermore, based on the quantile cointegration approach, the results indicate that the EKC hypothesis holds in 12 out of the 15 countries. However, even for these three countries, the EKC hypothesis seems to hold at certain quantiles.

Data files

Simulation files