Abstract: 

A news media index is usually constructed from a calibrated keyword set that is optimized on a topical scope; It cannot easily handle topic refinement, or non-English news articles, which greatly limit the application. Our paper overcomes topical scope restriction and language barrier; We filter news articles published around events that trigger media coverage on a topical scope, and back-out journalists' implied keywords. The event choice facilitates topical scope refinement, and the implied keywords originate from the same-language news articles they are meant to track. Using Engle et al. (2020) climate-document indices as benchmarks, our climate-event indices (physical risk and policy uncertainty) are more responsive in equity and futures market tests. We also use drought and flood events to construct physical tail-risk indices. Utilizing IPCC events, we extract Chinese keywords to construct climate indices for China, which we compare against those using Chinese-translated IPCC reports. The event indices exhibit greater explanatory power on commodity futures trading, and this holds against document indices translated using Baidu, Google and ChatGPT. We also use IPCC events to extract keywords in Japanese, Korean and French, and construct translation-free climate news indices in major languages.

Biography:

Prof. Michael T. Chng is a Professor of Finance at the International Business School Suzhou (IBSS) at Xi’an Jiaotong-Liverpool University (XJTLU). During his two terms as Associate Dean for Research, IBSS became the youngest business school in history to achieve AACSB-EQUIS-AMBA triple-crown accreditation. Michael earned his PhD from the Department of Finance, University of Melbourne. He has published in the Journal of Econometrics, Journal of Banking and Finance, Journal of Business Finance and Accounting and Journal of Futures Markets, a number of which are single-authored papers. Michael has received best paper awards and honorary mentions from various conferences, including AsianFA, Australasian Banking and Finance Meeting, FMA U.S, Asia-Pacific Association of Derivatives Meeting, International Conference on Futures and Derivative Markets, and Securities and Financial Markets Conference. His current research includes arbitrage mechanisms in financial markets, and machine-learning applications to empirical asset pricing.