Author:Itay Goldstein、Chester S. Spatt、Mao Ye.

原文鏈接: https://academic.oup.com/rfs/article/34/7/3213/6210658

Abstract:?data is revolutionizing the finance industry and has the potential to significantly shape future research in finance. This special issue contains papers following the 2019 NBER-RFS Conference on Big Data. In this introduction to the special issue, we define the “big data”phenomenon as a combination of three features: large size, high dimension, and complex structure. Using the papers in the special issue, we discuss how new research builds on these features to push the frontier on fundamental questions across areas in finance—including corporate finance, market microstructure, and asset pricing. Finally, we offer some thoughts for future research directions. (JEL G12, G14, G3)

The RFS has the tradition of encouraging scholars to pursue risky projects that have the potential to push the frontiers of research in finance

It is fairly clear that a definition of big data in finance research should be different from ones that are used in engineering and statistics. Researchers in these disciplines focus on providing facilities and tools to capture, curate,manage, and process data. Financial economists, on the other hand, focus on applying these tools to address interesting economic questions

big data in finance research: large size, high dimension, and complex structure

Unstructured data create value if they can measure economic activities that cannot be captured using structured data.

Corporate culture is important because it is perceived to be a key factor behind many business successes and failures (Graham et al.2018), and it is thought to be able to solve problems that cannot be regulated properly ex ante (Guiso, Sapienza, and Zingales 2015)

The “bag of words” approach is good at predicting the tone of a document by counting positive or negative words, but it is hard to capture important semantic information in an earnings call. The authors provide a method to decompose corporate culture onto a five-dimensional space of innovation, integrity, quality, respect, and teamwork, which are the five most-often mentioned values by the S&P 500 firms (see Guiso, Sapienza, and Zingales 2015)

The question becomes extremely important because algorithms and data increasingly became a major resource for the economy, particularly for finance. Back in 2017, the Economist published a story titled “The World's Most Valuable Resource Is No Longer Oil, but Data,”

Machine learning is one way to describe the world, and we also need theory to explain the world.

NBER has posted videotaped lectures for researchers in economics and finance on the application process for such free resources on the webpage for the 2018 Summer Conference on Big Data（http://www2.nber.org/si2018_video/bigdatafinancialecon/.）