nep-fmk 2023-07-24 papers

on Financial Markets

Issue of 2023‒07‒24
seven papers chosen by

Avoiding Idiosyncratic Volatility: Flow Sensitivity to Individual Stock Returns By Marco Di Maggio; Francesco Franzoni; Shimon Kogan; Ran Xing
Bloated Disclosures: Can ChatGPT Help Investors Process Financial Information? By Alex Kim; Maximilian Muhn; Valeri Nikolaev
Strong vs. Stable: The Impact of ESG Ratings Momentum and their Volatility on the Cost of Equity Capital By Ian Berk; Massimo Guidolin; Monia Magnani
Efficient Solution of Portfolio Optimization Problems via Dimension Reduction and Sparsification By Cassidy K. Buhler; Hande Y. Benson
Bank Funding, SME lending and Risk Taking By Sander Lammers; Massimo Giuliodori; Robert Schmitz; Adam Elbourne
Aggregate Insider Trading and Stock Market Volatility in the UK By Guglielmo Maria Caporale; Kyriacos Kyriacou; Nicola Spagnolo
This study compares well-known tools of technical analysis (Moving Average Crossover MAC) with Machine Learning based strategies (LSTM and XGBoost) and Ensembled Machine Learning Strategies (LSTM ensembled with XGBoost and MAC). All models were compared to Buy and Hold benchmark and evaluated using Performance Metrics, that is Annualized Return Compounded, Maximum Drawdown, Maximum Loss Duration, and three types of Information Ratio. This research uses daily S&P 500 index data ranging from 2000 to 2023. Every strategy was optimized with novel walk forward approach consisting of numerous in sample and out of sample periods. MAC and best performing ML methods were subjected to sensitivity analysis. The results show that LSTM ensembled with XGBoost and MAC yields the most promising results in terms of risk-adjusted returns which suggest further research focused on ensembling of individual ML strategies. Finally, we show that classical methods of technical analysis (that is, MAC) are much less robust and indifferent to change in hyperparameters than machine learning based algorithms, especially LSTM. By Karol Chojnacki; Robert Ślepaczuk

Avoiding Idiosyncratic Volatility: Flow Sensitivity to Individual Stock Returns

By:	Marco Di Maggio; Francesco Franzoni; Shimon Kogan; Ran Xing
Abstract:	Despite positive and significant earnings announcement premia, we find that institutional investors reduce their exposure to stocks before earnings announcements. A novel result on the sensitivity of flows to individual stock returns provides a potential explanation. We show that extreme announcement returns for an individual holding lead to substantial outflows, controlling for overall performance, and they increase the probability of managers leaving the fund. Reducing the exposure to these stocks before the announcement mitigates the outflows. We build a model to describe and quantify this tradeoff. Overall, the paper identifies a new dimension of limits to arbitrage for institutions.
JEL:	G12 G23
Date:	2023–06
URL:	http://d.repec.org/n?u=RePEc:nbr:nberwo:31360&r=fmk

Bloated Disclosures: Can ChatGPT Help Investors Process Financial Information?

By:	Alex Kim; Maximilian Muhn; Valeri Nikolaev
Abstract:	Generative AI tools such as ChatGPT can fundamentally change the way investors process information. We probe the economic usefulness of these tools in summarizing complex corporate disclosures using the stock market as a laboratory. The unconstrained summaries are dramatically shorter, often by more than 70% compared to the originals, whereas their information content is amplified. When a document has a positive (negative) sentiment, its summary becomes more positive (negative). More importantly, the summaries are more effective at explaining stock market reactions to the disclosed information. Motivated by these findings, we propose a measure of information "bloat." We show that bloated disclosure is associated with adverse capital markets consequences, such as lower price efficiency and higher information asymmetry. Finally, we show that the model is effective at constructing targeted summaries that identify firms' (non-)financial performance and risks. Collectively, our results indicate that generative language modeling adds considerable value for investors with information processing constraints.
Date:	2023–06
URL:	http://d.repec.org/n?u=RePEc:arx:papers:2306.10224&r=fmk

Strong vs. Stable: The Impact of ESG Ratings Momentum and their Volatility on the Cost of Equity Capital

By:	Ian Berk; Massimo Guidolin; Monia Magnani
Abstract:	We test the performance of two ESG score-driven quantitative signals on a large, multi-national crosssection of European stock returns. In particular, we ask whether in the cross-section, the cost of equity capital is more strongly affected by the (upward) “slope” (identified as momentum over a period of time) of their ESG scores or by their “stability” (identified as the volatility of the scores over a period of time), measured around a given slope. We find that 1 month, short-term ESG momentum is priced in the cross-section of stock returns and that it lowers the ex-ante cost of capital (at the same time causing realised ex post average abnormal returns). Short-term ESG momentum may represent a novel, priced systematic risk factor. There is equally strong evidence that a ESG spread strategy that buys (sells) low (high) ESG score volatility stocks leads to a significant alpha and alters the ex-ante cost of capital. Both quantitative ESG signals lead to portfolio sorts and long-short strategies that increase the speed of improvement of the aggregate sustainability profile of the resulting portfolios with no costs in terms of average ESG scores or their stability
Keywords:	ESG ratings, ESG momentum, ESG score volatility, cross-sectional pricing, systematic risk factor.
JEL:	G11 G12 C59 G24
Date:	2023
URL:	http://d.repec.org/n?u=RePEc:baf:cbafwp:cbafwp23202&r=fmk

Efficient Solution of Portfolio Optimization Problems via Dimension Reduction and Sparsification

By:	Cassidy K. Buhler; Hande Y. Benson
Abstract:	The Markowitz mean-variance portfolio optimization model aims to balance expected return and risk when investing. However, there is a significant limitation when solving large portfolio optimization problems efficiently: the large and dense covariance matrix. Since portfolio performance can be potentially improved by considering a wider range of investments, it is imperative to be able to solve large portfolio optimization problems efficiently, typically in microseconds. We propose dimension reduction and increased sparsity as remedies for the covariance matrix. The size reduction is based on predictions from machine learning techniques and the solution to a linear programming problem. We find that using the efficient frontier from the linear formulation is much better at predicting the assets on the Markowitz efficient frontier, compared to the predictions from neural networks. Reducing the covariance matrix based on these predictions decreases both runtime and total iterations. We also present a technique to sparsify the covariance matrix such that it preserves positive semi-definiteness, which improves runtime per iteration. The methods we discuss all achieved similar portfolio expected risk and return as we would obtain from a full dense covariance matrix but with improved optimizer performance.
Date:	2023–06
URL:	http://d.repec.org/n?u=RePEc:arx:papers:2306.12639&r=fmk

Bank Funding, SME lending and Risk Taking

By:	Sander Lammers (CPB Netherlands Bureau for Economic Policy Analysis); Massimo Giuliodori (UVA); Robert Schmitz; Adam Elbourne (CPB Netherlands Bureau for Economic Policy Analysis)
Abstract:	European companies heavily rely on bank credit to finance their operations and investments. Therefore, it is crucial for banks to take risks on corporate loans, although excessive risk-taking can have negative consequences for banks. There are indications in the literature that the financing structure used by banks plays a role in determining the risks they take. However, the economic literature does not provide clear consensus on how different elements of a bank's financing structure are related to risk. In this exploratory study, we investigated this relationship specifically focusing on corporate loans. This contributes to a better understanding of which companies receive funding and how a bank's financing structure itself can become a risk, particularly when riskier companies face bankruptcy. The financing structure of banks primarily consists of equity (capital buffer), deposits (savings from households and businesses), market financing (funds raised from international investors), and interbank loans (loans between banks, including central bank loans). We analyzed the extent to which these individual financing elements contribute to the risks banks take on loans to companies.
JEL:	G21 G32 E52
Date:	2023–07
URL:	http://d.repec.org/n?u=RePEc:cpb:discus:447&r=fmk

Aggregate Insider Trading and Stock Market Volatility in the UK

By:	Guglielmo Maria Caporale; Kyriacos Kyriacou; Nicola Spagnolo
Abstract:	This paper examines the relationship between aggregate insider trading (AIT) and stock market volatility using monthly data on insider transactions by UK executives in public limited companies for the period January 2002 - December 2020. More specifically, a Vector Autoregression (VAR) model is estimated and Impulse Response analysis as well as Forecast Error Variance Decomposition are carried out. The main finding is that higher AIT (more specifically, insider purchases) leads to a short-run increase in stock market volatility; this can be attributed to a combination of insiders manipulating the timing and content of the information they release and the revelation of new economy-wide information to the market. The UK being a well-regulated market, it is plausible that the main driver of the increase in stock market volatility should be the information effect. These results are shown to be robust to using alternative (direct) measures of AIT.
Keywords:	aggregate insider trading, stock market volatility, VAR, impulse responses
JEL:	C22 G14
Date:	2023
URL:	http://d.repec.org/n?u=RePEc:ces:ceswps:_10511&r=fmk

This study compares well-known tools of technical analysis (Moving Average Crossover MAC) with Machine Learning based strategies (LSTM and XGBoost) and Ensembled Machine Learning Strategies (LSTM ensembled with XGBoost and MAC). All models were compared to Buy and Hold benchmark and evaluated using Performance Metrics, that is Annualized Return Compounded, Maximum Drawdown, Maximum Loss Duration, and three types of Information Ratio. This research uses daily S&P 500 index data ranging from 2000 to 2023. Every strategy was optimized with novel walk forward approach consisting of numerous in sample and out of sample periods. MAC and best performing ML methods were subjected to sensitivity analysis. The results show that LSTM ensembled with XGBoost and MAC yields the most promising results in terms of risk-adjusted returns which suggest further research focused on ensembling of individual ML strategies. Finally, we show that classical methods of technical analysis (that is, MAC) are much less robust and indifferent to change in hyperparameters than machine learning based algorithms, especially LSTM.

By:	Karol Chojnacki (University of Warsaw, Faculty of Economic Sciences); Robert Ślepaczuk (University of Warsaw, Quantitative Finance Research Group, Department of Quantitative Finance, Faculty of Economic Sciences)
Keywords:	Algorithmic Investment Strategies, Machine Learning, Recurrent Neural Networks, Long Short-Term Memory, XGBoost, Walk Forward Optimization, Trading algorithms, Technical Analysis Indicators
JEL:	C4 C14 C45 C53 C58 G13
Date:	2023
URL:	http://d.repec.org/n?u=RePEc:war:wpaper:2023-15&r=fmk

General information on the NEP project can be found at https://nep.repec.org. For comments please write to the director of NEP, Marco Novarese at <director@nep.repec.org>. Put “NEP” in the subject, otherwise your mail may be rejected.

NEP’s infrastructure is sponsored by the School of Economics and Finance of Massey University in New Zealand.