We add to the emerging literature on empirical asset pricing in Chinese stock market by building and analyzing a comprehensive set of return prediction factors using various machine learning algorithms. Contrasting previous studies for US market, liquidity emerges as most important predictor, leading us closely examine impact transaction costs. The retail investors' dominating presence positively affects short-term predictability, particularly small stocks. Another feature that distinguishes...

Abstract This paper performs specification analysis on the term structure of variance swap rates S&P 500 index and studies optimal investment decision swaps stock index. The identifies 2 stochastic risk factors, which govern short long end variation, respectively. highly negative estimate for market price makes it an investor to take positions in a short-term contract, long-term

Disclosure of climate-related financial risks greatly helps investors assess companies’ preparedness for climate change. Voluntary disclosures such as those based on the recommendations Task Force Climate-related Financial Disclosures (TCFD) are being hailed an effective measure better risk management. We ask whether this expectation is justified. do so by training ClimateBERT, a deep neural language model fine-tuned BERT. In analyzing TCFD-supporting firms, ClimateBERT comes to sobering...

In response to the Russia-Ukraine war, stocks more exposed regulatory risks of transition a low-carbon economy performed better, suggesting that investors expect an overall slowdown in this transition. These stock price effects were particularly strong US. Europe, less pronounced or even opposite. Arguably, market participants initially expected stronger policy responses supporting European renewable energy sources. Investors perceived both REPowerEU plan and US Inflation Reduction Act boost...

Abstract We use BERT, an AI-based algorithm for language understanding, to quantify regulatory climate risk disclosures and analyze their impact on the term structure in credit default swap (CDS) market. Risk can either increase or decrease CDS spreads, depending whether disclosure reveals new risks reduces uncertainty. Training BERT differentiate between transition physical risks, we find that disclosing increases spreads after Paris Climate Agreement of 2015, while decreases spreads. In...

Over the recent years, large pretrained language models (LM) have revolutionized field of natural processing (NLP). However, while pretraining on general has been shown to work very well for common language, it observed that niche poses problems. In particular, climate-related texts include specific LMs can not represent accurately. We argue this shortcoming today's limits applicability modern NLP broad text texts. As a remedy, we propose ClimateBert, transformer-based model is further over...

Abstract Large Language Models have made remarkable progress in question-answering tasks, but challenges like hallucination and outdated information persist. These issues are especially critical domains climate change, where timely access to reliable is vital. One solution granting these models external, scientifically accurate sources enhance their knowledge reliability. Here, we GPT-4 by providing the Sixth Assessment Report of Intergovernmental Panel on Climate Change (IPCC AR6), most...

In this study, we explore the susceptibility of financial sentiment analysis to adversarial attacks that manipulate texts. With rise AI readership in sector, companies are adapting their language and disclosures fit processing better, leading concerns about potential for manipulation. finance literature, keyword-based methods, such as dictionaries, still widely used due perceived transparency. However, our research demonstrates vulnerability approaches by successfully generating using...

Navigating the complex landscape of corporate climate disclosures and their real impacts is crucial for managing climate-related financial risks. However, current oftentimes suffer from imprecision, inaccuracy, greenwashing. We introduce ClimateBert CTI, a deep learning algorithm, to identify cheap talk in MSCI World index firms' annual reports. find that only targeted engagement associated with less talk. Voluntary are more Moreover, correlates increased negative news coverage higher...

In this paper, we survey a large sample of Swiss households to measure sustainable finance literacy, which define as the knowledge and skill identifying assessing financial products according their reported sustainability-related characteristics. To end, use multiple-choice questions. Furthermore, private investors' level awareness about using open-ended We find that households, are generally highly financially literate by international standards, exhibit low levels literacy compared current...

We identify and characterize a class of term structure models where bond yields are quadratic functions the state vector.We label this aim to lay solid theoretical foundation for its future empirical application.We consider asset pricing in general derivative particular under class.We provide two transform methods wide variety fixed income derivatives closed or semi-closed form.We further illustrate how model can be applied more settings.

This paper is an interview with a Large Language Model (LLM), namely GPT-3, on the issues of climate change. The should give some insights into current capabilities these large models which are deep neural networks generally more than 100 billion parameters. In particular, it shows how eloquent and convincing answers such LLMs can be. However, be noted that suffer from hallucination their responses may not grounded facts. These deficiencies offer interesting avenue for future research.

Environmental, social, and governance (ESG) criteria take a central role in fostering sustainable development economies. This paper introduces class of novel Natural Language Processing (NLP) models to assess corporate disclosures the ESG subdomains. Using over 13.8 million texts from reports news, specific E, S, G were pretrained. Additionally, three 2k datasets developed classify ESG-related texts. The effectively explain variations ratings, showcasing robust method for enhancing...

Journal Article Learning and Asset Prices Under Ambiguous Information Get access Markus Leippold, Leippold Search for other works by this author on: Oxford Academic Google Scholar Fabio Trojani, Trojani Paolo Vanini The Review of Financial Studies, Volume 21, Issue 6, November 2008, Pages 2565–2597, https://doi.org/10.1093/rfs/hhm035 Published: 12 September 2008

We introduce CLIMATE-FEVER, a new publicly available dataset for verification of climate change-related claims. By providing the research community, we aim to facilitate and encourage work on improving algorithms retrieving evidential support climate-specific claims, addressing underlying language understanding challenges, ultimately help alleviate impact misinformation change. adapt methodology FEVER [1], largest artificially designed real-life claims collected from Internet. While during...

Investors and regulators require reliable estimates of physical climate risks for decision-making. While assessing these is challenging, several commercial data providers academics have started to develop firm-level risk scores. We compare six find a substantial divergence between scores, also among those based on similar methodologies. show how this could cause problems when testing whether financial markets are pricing risks. Hence, may not adequately account the exposure corporations...

This paper introduces the concept of sustainable finance literacy. We survey a large sample Swiss households to measure To this end, we use multiple-choice questions and novel approach based on open-ended natural language processing. find that households, which are generally highly financially literate by international standards, exhibit low levels financial Moreover, despite its level, knowledge about is significant factor in ownership products. Therefore, our results show an urgent need...

Large language models (LLMs) have significantly transformed the landscape of artificial intelligence by demonstrating their ability to generate human-like text across diverse topics. However, despite impressive capabilities, LLMs lack recent information and often employ imprecise language, which can be detrimental in domains where accuracy is crucial, such as climate change. In this study, we make use ideas harness potential viewing them agents that access multiple sources, including...

We consider the design and estimation of quadratic term structure models. start with a list stylized facts on interest rates rate derivatives, classified into three layers: (1) general statistical properties, (2) forecasting relations, (3) conditional dynamics. then investigate implications each layer property model strive to establish mapping between evidence structures. calibrate two-factor that approximates these layers properties well, show flexible specification for market price risk is...

To simultaneously account for the properties of interest-rate term structure and foreign exchange rates within one arbitrage-free framework, we propose a class multi-currency quadratic models (MCQM) with an (m + n) factor in pricing kernel each economy. The m factors model interest rates. n capture portion rate movement that is independent structure. Our modeling framework represents first literature not only explicitly allows currency movement, but also guarantees internal consistency...

Abstract We study the impact of economic policy uncertainty on term structure nominal interest rates. In a general equilibrium model populated by an averse agent, we show that political not only affects yield curve and corresponding volatility but also bond risk premia carry premium for uncertainty. Our simultaneously captures both shape hump volatilities, stylized feature is hard to match with theoretical model. gives rise set testable predictions which find strong support in data: Higher...