Exploring Accounting Research Topic Evolution: An Unsupervised Machine Learning Approach
Journal of International Accounting Research,
This study explores the evolution of accounting research by utilizing an unsupervised machine learning approach. We aim to identify the latent topics of accounting from the 1980s up to 2018, the dynamics and emerging topics of accounting research, and the economic reasons behind those changes. First, based on 23,220 articles from 46 accounting journals, we identify 55 topics using the latent Dirichlet allocation model. To illustrate the connection between topics, we use HistCite to generate a citation map along a timeline. The citation clusters demonstrate the “tribalism” phenomenon in accounting research. We then implement the dynamic topic model to reveal the dynamics of topics to show changes in accounting research. The emerging research trends are identified from the topic analytics. We further explore the economic reasons and in-depth insights into the topic evolution, indicating the economic development embeddedness nature of accounting research.
"Let Me Get Back to You" — A Machine Learning Approach to Measuring NonAnswers
Using a supervised machine learning framework on a large training set of questions and answers, we identify 1,364 trigrams that signal nonanswers in earnings call questions and answers (Q&A). We show that this glossary has economic relevance by applying it to contemporaneous stock market reactions after earnings calls. Our findings suggest that obstructing the flow of information leads to significantly lower cumulative abnormal stock returns and higher implied volatility. As both our method and glossary are free of financial context, we believe that the measure is applicable to other fields with a Q&A setup outside the contextual domain of financial earnings conference calls.
23.05.2023 • 14/2023
Analysis of earnings calls: Blathering managers harm their company
If a senior executive refuses to give information to professional investors, the company's stock market value drops afterwards. This is shown in a study by the Halle Institute for Economic Research (IWH) after evaluating
1.2 million answers from tele-phone conferences.
Read press release
DPE Course Programme Archive
DPE Course Programme Archive 2023 2022 2021 2020 2019...
Research Profiles of the IWH Departments All doctoral students are allocated to one...
On Modeling IPO Failure Risk
This paper offers a novel framework, combining firm operational risk, IPO pricing risk, and market risk, to model IPO failure risk. By analyzing nearly a thousand variables, we observe that prior IPO failure risk models have suffered from a major missing-variable problem. Evidence reveals several key new firm-level determinants, e.g., the volatility operating performance, the size of its accounts payable, pretax income to common equity, total short-term debt, and a few macroeconomic variables such as treasury bill rate, and book-to-market of the DJIA index. These findings have major economic implications. The total value loss from not predicting the imminent failure of an IPO is significantly lower with this proposed model compared to other established models. The IPO investors could have saved around $18billion over the period between 1994 and 2016 by using this model.
Evolvement of China-related Topics in Academic Accounting Research: Machine Learning Evidence
China Accounting and Finance Review,
This study employs an unsupervised machine learning approach to explore the evolution of accounting research. We are particularly interested in exploring why international researchers and audiences are interested in China-related issues; what kinds of research topics related to China are mainly investigated in globally recognised journals; and what patterns and emerging topics can be explored by comprehensively analysing a big sample. Using a training sample of 23,220 articles from 46 accounting journals over the period 1980 to 2018, we first identify the optimal number of accounting research topics; the dynamic patterns of these accounting research topics are explored on the basis of 46 accounting journals to show changes in the focus of accounting research. Further, we collect articles related to Chinese accounting research from 18 accounting journals, eight finance journals, and eight management journals over the period 1980 to 2018. We objectively identify China-related accounting research topics and map them to the stages of China’s economic development. We attempt to identify the China-related issues global researchers are interested in and whether accounting research reflects the economic context. We use HistCite TM to generate a citation map along a timeline to illustrate the connections between topics. The citation clusters demonstrate “tribalism” phenomena in accounting research. The topics related to Chinese accounting research conducted by international accounting researchers reveal that accounting changes mirror economic reforms. Our findings indicate that accounting research is embedded in the economic context.
Why are some Chinese Firms Failing in the US Capital Markets? A Machine Learning Approach
Pacific-Basin Finance Journal,
We study the market performance of Chinese companies listed in the U.S. stock exchanges using machine learning methods. Predicting the market performance of U.S. listed Chinese firms is a challenging task due to the scarcity of data and the large set of unknown predictors involved in the process. We examine the market performance from three different angles: the underpricing (or short-term market phenomena), the post-issuance stock underperformance (or long-term market phenomena), and the regulatory delistings (IPO failure risk). Using machine learning techniques that can better handle various data problems, we improve on the predictive power of traditional estimations, such as OLS and logit. Our predictive model highlights some novel findings: failed Chinese companies have chosen unreliable U.S. intermediaries when going public, and they tend to suffer from more severe owners-related agency problems.
Does Machine Learning Help us Predict Banking Crises?
Journal of Financial Stability,
This paper compares the out-of-sample predictive performance of different early warning models for systemic banking crises using a sample of advanced economies covering the past 45 years. We compare a benchmark logit approach to several machine learning approaches recently proposed in the literature. We find that while machine learning methods often attain a very high in-sample fit, they are outperformed by the logit approach in recursive out-of-sample evaluations. This result is robust to the choice of performance metric, crisis definition, preference parameter, and sample length, as well as to using different sets of variables and data transformations. Thus, our paper suggests that further enhancements to machine learning early warning models are needed before they are able to offer a substantial value-added for predicting systemic banking crises. Conventional logit models appear to use the available information already fairly efficiently, and would for instance have been able to predict the 2007/2008 financial crisis out-of-sample for many countries. In line with economic intuition, these models identify credit expansions, asset price booms and external imbalances as key predictors of systemic banking crises.
Predicting Free-riding in a Public Goods Game – Analysis of Content and Dynamic Facial Expressions in Face-to-Face Communication
IWH Discussion Papers,
This paper illustrates how audio-visual data from pre-play face-to-face communication can be used to identify groups which contain free-riders in a public goods experiment. It focuses on two channels over which face-to-face communication influences contributions to a public good. Firstly, the contents of the face-to-face communication are investigated by categorising specific strategic information and using simple meta-data. Secondly, a machine-learning approach to analyse facial expressions of the subjects during their communications is implemented. These approaches constitute the first of their kind, analysing content and facial expressions in face-to-face communication aiming to predict the behaviour of the subjects in a public goods game. The analysis shows that verbally mentioning to fully contribute to the public good until the very end and communicating through facial clues reduce the commonly observed end-game behaviour. The length of the face-to-face communication quantified in number of words is further a good measure to predict cooperation behaviour towards the end of the game. The obtained findings provide first insights how a priori available information can be utilised to predict free-riding behaviour in public goods games.