Sustainability in Malaysian Sukuk Issuers: Financial vs. Non-financial Firms via Machine Learning

It is an empirical question whether the use of derivatives hedging among firms actually contributes to enhancing firm performances. Despite the increasing use of derivatives by non-financial firms, existing literature still debates about their effect, especially in countries with peculiar corporate governance mechanisms. By using a sample of non-financial Italian firms listed from 2007 to 2018, this paper investigates if the use of several types (currency, interest rate, and commodity) of financial derivatives can affect the value of a company. For measuring the impact of the derivatives and in order to address any possible endogeneity problem, besides using the conventional methodologies applied by previous literature (fixed-effect regression models and system GMM estimators), we run a random forest model, a machine learning technique not yet applied before in this field, and calculate the relative importance of each independent and control variable. Differently from other European countries, findings show that the use of derivatives does not affect the firm value in the Italian market. Therefore, our results confirm the role of corporate governance mechanisms on the relationship between firm value and the use of derivatives and that their impact is country-specific.

Examining the asymmetric impact of macroeconomic policy in the UAE: Evidence from quartile impulse responses and machine learning

AI (artificial intelligence) is a significant technological advancement that has everyone buzzing about its incredible potential. The current research study evaluates the influence of supervised artificial intelligence techniques, i.e., machine learning techniques on the nonfinancial firms of Pakistan and focuses on the practical application of AI techniques for the accurate prediction of corporate risks which in turn will lead to the automation of corporate risk management. So, in this study, we used financial ratios for accurate risk assessment and for the automation of corporate risk management by developing machine learning algorithms using techniques, namely, random forest, decision tree, naïve Bayes, and KNN. A secondary data collection technique will be used. For this purpose, we collected annual data of nonfinancial companies in Pakistan for the period ranging from 2006 to 2020, and the data are analyzed and tested through Python software. Our results prove that AI techniques can accurately predict risk with minimum error values, and among all the techniques used, the random forest technique outperforms as compared to the rest of the techniques.

PurposeThe global sukuk market has seen widespread innovations in the last couple of decades, which helped sukuk develop into one of the most acceptable Islamic instruments for raising finance. According to the State of the Global Islamic Economy Report (2018–19), United Arab Emirates (UAE) is ranked second among Islamic economies and Nasdaq Dubai is credited to be the leading international center for sukuk listings (Thomson Reuters, 2018). However, there has been limited research studies on this financing option within the region. To the best of the authors’ knowledge, this study is the first to focus on the role of signaling theory driving the financing choice for listed entities in the UAE. The paper aims to make a significant contribution in light of the recent expansion of sukuk issuances and fills the lacuna in research carried out in the UAE bond market.Design/methodology/approachThis study empirically tested the hypotheses on a data set that covered a sample of 1,354 bond issuances over the period 2008–2019. The authors used logistic regression to distinguish between the issuance of sukuk versus conventional bond. Sukuk structuration leads to information asymmetry that prompts firms to send signals to the capital market. Information asymmetry has been studied in terms of issue-specific (maturity and issue size) and issuer-specific (firm size, growth, profitability, leverage) variables. Two control variables were included to capture the years under study and the macroeconomic effects of economic slowdown.FindingsThe banking sector accounted for 93% of bond issuances but contributed only 63% of the bond market in the UAE in terms of issue size. The data evidenced that non-banking sukuk issuances expanded over the years, with participation from sectors like real estate, oil and gas, logistics and utilities and contributed 50% of issuances in the UAE sukuk market. Typically, firms with smaller asset sizes and higher financing requirements were found to favour sukuk. The banking sector revealed irrelevance of information asymmetry, as Islamic Banks were mandated to issue sukuk. Non-financial firms with high profits and high debts were prompted to prefer conventional bonds, in line with the adverse selection mechanism.Originality/valueAlthough UAE’s sukuk market has existed for more than a decade, scant research has been carried out. Few studies exist for the GCC region that either concentrated on stock market reactions to issuances of Islamic versus conventional bonds or studied capital market characteristics of non-financial entities alone. This is the first study to focus on signaling theory and information asymmetry playing a role in the capital structure of all listed firms (banking and non-banking) issuing bonds in the UAE.

The capital market of Saudi Arabia has experienced tremendous growth in Islamic finance tools like Sukuk under its economic development plan, named Saudi Vision 2030. It is essential for the capital market to evaluate the credit risk associated with Sukuk issuers for proper functioning. Generally, credit risk evaluation is performed using linear statistical and judgmental methods. However, it is difficult to evaluate credit risk using traditional methods because of the non-linear relationship between variables. In this study, a machine learning approach has been proposed for credit risk evaluation for Sukuk issuers. Using supervised machine learning techniques like Logistic Regression, Random Forest, Support Vector Machines, and Gradient Boosting, the study has evaluated various financial and macroeconomic variables associated with Sukuk issuers. The results indicate that the machine learning approach is much better for credit risk evaluation compared to traditional methods. The findings of this study will help implement Saudi Vision 2030 because it will highlight the use of artificial intelligence for credit risk evaluation in Islamic capital markets.

PurposeThis study examines the influence of corporate governance indicators (CGIs) on the textual tone of nonfinancial firms in a developing economy.Design/methodology/approachThe data from 1,250 annual reports of listed nonfinancial firms in Pakistan are collected for 10 years. The narrative disclosure tone (NDT) is derived using the sentiment analysis of annual reports, resulting in six distinct NDT scores. The CGIs data are also extracted from the annual reports. The fixed effects model is used as the primary analytical tool, supplemented by machine learning-based linear regression. System GMM and two-stage least squares regressions are employed for robustness checks.FindingsThe findings reveal that most CGIs significantly influence all six NDTs. These results align with the existing theoretical literature, except those related to audit committee independence and gender diversity.Research limitations/implicationsThe study is limited to the use of annual reports as a source of narrative disclosures. Future research might employ other sources, such as earning press releases and social media.Practical implicationsWithin the unique regulatory environment of Pakistan, the study offers insights for regulators to enhance the efficacy of independent directors, discourage concentrated ownership and promote the inclusion of women in board subcommittees to establish the authenticity of textual disclosures.Originality/valueThe study adds to the limited literature on the determinants of NDT. It underscores the importance of understanding textual tone for informed investor decision-making and restoring investor confidence. Moreover, it contributes by focusing on six NDTs and exploring the interplay between CGIs and textual tone.

Financial Innovation in the 21st Century: Evidence from U.S. Patents

We develop a unique dataset of 24 thousand U.S. finance patents granted over last two decades to explore the evolution and production of financial innovation. We use machine learning to identify the financial patents and extensively audit the results to ensure their reasonableness. We find that patented financial innovation is substantial and economically important, with the number of annual grants expanding from a few dozen in the 1990s to over 2000 in the 2010s. The subject matter of financial patents has changed, consistent with the industry’s shift in revenue and value-added towards household investors and borrowers. The surge in financial patenting was driven by information technology firms and others outside of financial sector, which collectively accounted for 69% of the awards. The location of innovation has shifted, with banks moving this activity from regions with tight financial regulation to more permissive ones. High-tech regions have attracted financial innovation by payments, IT, and other non-financial firms. Turning to the source of these ideas, while academic knowledge remained associated with more valuable patents, citations in finance patents to academic papers, especially in those by banks, fell sharply. Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.

Financial Innovation in the 21st Century: Evidence from U.S. Patents

This study examines the impact of Artificial Intelligence (AI)-enhanced Environmental, Social, and Governance (ESG) reporting on firm valuation in emerging markets. It aims to explore how AI integration enhances the interpretability and predictive accuracy of ESG metrics in shaping market perceptions and investor decisions. This study employs a panel dataset from 2018 to 2024, analysing publicly listed non-financial firms across five major sectors: manufacturing, energy, telecommunications, consumer goods, and industrials. This study contributed by employing AI-powered multimodal analysis with conventional ESG scoring methods and integrating Fixed-Effects Regression with machine learning (ML) algorithms including Extreme Gradient Boosting (XGBoost) and Random Forest to identify complex, non-linear relationships within ESG data and firm valuation. The results show empirical evidence that integrating ML enhances the explanatory power of ESG data. Findings indicate that ESG performance is positively correlated with higher market valuations, particularly in Environmental and Social dimensions. Governance metrics are more inconsistent, which may be due to heterogeneity in governance practices, regulatory enforcement and the challenges of quantifying governance quality beyond compliance indicators across the focus emerging markets. Firms identified in ESG controversies tend to face valuation penalties, which stresses market sensitivity to reputational risks. ML algorithms outperform conventional techniques in predictive accuracy, revealing complex, non-linear interactions within ESG data. This study contributes to both the academic literature and practice showing how next-generation ESG reporting can robust valuation models, address limitations of conventional ESG scoring, and ensure a reliable outlook for investors and policymakers of industries in emerging markets.

This study investigates the prediction of multi-class S&P corporate credit ratings for European non-financial firms from 2010 to 2024 using a machine learning framework grounded in financial fundamentals. To ensure robustness and generalizability, the analysis excluded the Year variable, which was identified as a source of data leakage. After this correction, non-linear ensemble models demonstrated a clear advantage over linear baselines. The top-performing Random Forest model achieved a weighted F1-score of approximately 0.60, more than doubling the performance of the Logistic Regression benchmark used as a baseline (0.26), with most misclassifications concentrated in adjacent rating categories. This indicates that while precise distinctions remain challenging, the models capture overall credit quality effectively. A SHAP (SHapley Additive exPlanations) analysis confirmed that predictions are driven by financially intuitive features, particularly long-term profitability (Retained_Earnings_to_Assets), firm size (Market Capitalization), and core earnings measures. These findings align with established credit risk theory, suggesting that a substantial portion of agency ratings is systematically associated with publicly available financial data. The research contributes a transparent and academically sound benchmark for quantitative credit risk assessment. While the results underscore the continuing importance of qualitative and forward-looking judgment applied by credit rating agencies, the study demonstrates that machine learning can provide a valuable, interpretable complement for credit monitoring, screening, and early-warning applications.

PurposeThis study attempts to identify fundamental determinants of bond ratings for non-financial and financial firms. Further the study aims to develop a parsimonious bond rating model and compare its efficacy across statistical and range of machine learning methods in the Indian context. The study is motivated by the insufficiency of prior work in the Indian context.Design/methodology/approachThe authors identify the critical determinants of non-financial and financial firms using multinomial logistic regression. Various machine learning and statistical methods are employed to identify the optimal bond rating prediction model. The data cover 8,346 bond issues from 2009 to 2019.FindingsThe authors find that industry concentration, sales, operating leverage, operating efficiency, profitability, solvency, strategic ownership, age, firm size and firm value play an important role in rating non-financial firms. Operating efficiency, profitability, strategic ownership and size are also relevant for financial firms besides additional determinants related to the capital adequacy, asset quality, management efficiency, earnings quality and liquidity (CAMEL) approach. The authors find that random forest outperforms logit and other machine learning methods with an accuracy rate of 92 and 91% for non-financial and financial firms.Practical implicationsThe study identifies important determinants of bond ratings for both non-financial and financial firms. The study interalia finds that the random forest technique is the most appropriate method for bond ratings predictions in India.Social implicationsBetter bond ratings may mitigate corporate defaults.Originality/valueUnlike prior literature, the study identifies determinants of bond ratings for both non-financial and financial firms. The study also experiments with modern machine learning techniques besides the traditional statistical approach for model building in case of relatively under researched market.

This paper constructs a leading macroeconomic indicator from microeconomic data using recent machine learning techniques. Using tree-based methods, we estimate probabilities of default for publicly traded non-financial firms in the United States. We then use the cross-section of out-of-sample predicted default probabilities to construct a leading indicator of non-financial corporate health. The index predicts real economic outcomes such as GDP growth and employment up to eight quarters ahead. Impulse responses validate the interpretation of the index as a measure of financial stress.

Purpose : The theories of dividend decision have disentangled the firms’ critical drivers of the dividend announcement, and their performances are empirically evaluated by employing ordinary least squares (OLS). However, after more than half a century of research, the debate over the determinants of dividend policy in firms is inconclusive. Therefore, the current study attempted to contribute to the literature by exploring new insights into the dividend decisions of Indian firms by employing machine learning. Methodology : This study is based on secondary data, and empirical analysis has used a novel dataset of 919 listed Indian nonfinancial firms from 1999–2019. The study utilized the least absolute shrinkage and selection operator and logistic regression methodologies. Findings : The findings revealed that the idiosyncratic variables are critically significant for dividend announcements by Indian firms. The results demonstrated that large, profitable, liquid, and firms with high market share were more likely to announce dividends in India than small, loss-making, illiquid, and low-market share firms. The direct relationship between Tobin’s Q and the likelihood of paying dividends is a new insight into the dividend decision for Indian firms. Practical Implications : The results will guide the dividend seeker investors to hold the shares of a high market share firm to receive the expected dividend. Originality/Value : This current study extended the literature by studying the dividend decisions of Indian firms by employing the machine learning methodology.

We compare statistical models usually employed in credit risk forecasting with machine learning algorithms (ML). Using a large dataset which includes financial ratios and credit behavioral indicators for about 300,000 Italian non-financial firms from 2011 to 2017, we show that training the models on financial statement data only, ML models record a significant improvement in discriminatory power and precision with respect to statistical models; however, this improvement is less pronounced when we enlarge the training dataset to include also credit behavioral data