With the rapid development of the global economy, crude oil is becoming more and more prominent in terms of national stability. However, oil prices dramatically fluctuate during emergencies. Meanwhile, network search data have been widely used for prediction during the era of big data. Herein, a suggestion is introduced for improving the traditional case analysis. An autoregressive distributed lag model is established, considering emergency and network search data. Moreover, a network attention index of specific emergencies is used to explain fluctuations of the oil price and the influence of this attention is analyzed. Results show: 1) major emergencies have a significant short-term impact on the international oil market and a remarkable influence on the cumulative abnormal return of an event window, and 2) market attention can aggravate fluctuations of oil prices. It is found that the individual network attention paid to each of four emergencies has a significant impact on oil prices. The network attention related to Hurricane Katrina and the Libyan war has positive effects on oil prices. However, the effects of network attention paid to the subprime crisis and the Mexico oil spill of 2010 are negative. The attention paid to the subprime crisis has both the greatest and the longest lasting impact.
Keywords: case analysis; crude oil price fluctuations; emergencies; network search data.
© 2020 The Authors. Published by Wiley‐VCH GmbH.