Search results

Public debt forecasts represent a key policy issue. Many methodologies have been employed to predict debt sustainability, including dynamic stochastic general equilibrium models, the stock flow consistent method, the structural vector autoregressive model and, more recently, the neuro-fuzzy method. Despite their widespread application in the empirical literature, all of these approaches exhibit shortcomings that limit their utility. The present research adopts a different approach to public debt forecasts, that is, the random forest, an ensemble of machine learning.

Using quarterly observations over the period 2000–2021, the present research tests the reliability of the random forest technique for forecasting the Italian public debt.

The results show the large predictive power of this method to forecast debt-to-GDP fluctuations, with no need to model the underlying structure of the economy.

Compared to other methodologies, the random forest method has a predictive capacity that is granted by the algorithm itself. The use of repeated learning, training and validation stages provides well-defined parameters that are not conditional to strong theoretical restrictions This allows to overcome the shortcomings arising from the traditional techniques which are generally adopted in the empirical literature to forecast public debt.

The purpose of our study is to study the impact of chief executive officer (CEO) optimism and narcissism on the company's capital structure.

(1) The authors opt for regression, machine learning and text analysis to explore the impact of narcissism and optimism on the capital structure. (2) We analyze CEO interviews and employ three methods to evaluate narcissism: the dictionary proposed by Anglin, which enabled us to assess the following components: authority, superiority, vanity and exhibitionism; count of first-person singular and plural pronouns and count of CEO photos displayed. Following this approach, we were able to make a more thorough assessment of corporate narcissism. (3) Latent Dirichlet allocation (LDA) technique helped to find the differences in the corporate rhetoric of narcissistic and non-narcissistic CEOs and to find differences between the topics of interviews and letters provided by narcissistic and non-narcissistic CEOs.

Our research demonstrates that narcissism has a slight and nonlinear impact on capital structure. However, our findings suggest that there is an impact of pessimism and uncertainty under pandemic conditions when managers predicted doom and completely changed their strategies. We applied various approaches to estimate the gender distribution of CEOs and found that the median values of optimism and narcissism do not depend on sex. Using LDA, we examined the content and key topics of CEO interviews, defined as positive and negative. There are some differences in the topics: narcissistic CEOs are more likely to speak about long-term goals, projects and problems; they often talk about their brand and business processes.

First, we examine the COVID-19 pandemic period and evaluate how CEO optimism and pessimism affect their financial decisions under specific external conditions. The pandemic forced companies to shift the way they worked: either to switch to the remote work model or to interrupt operations; to lose or, on the contrary, attract clients. In addition, during this period, corporate management can have a different outlook on their company’s financial performance and goals. The LDA technique helped to find the differences in the corporate rhetoric of narcissistic and non-narcissistic CEOs. Second, we use three methods to evaluate narcissism. Third, the research is based on a set of advanced methods: machine learning techniques (random forest to reveal a nonlinear impact of CEO optimism and narcissism on capital structure).

This paper aims to study the impact of geopolitical uncertainty on corporate cost of debt and the moderating role of information asymmetry between creditors and borrowing firms.

This study uses 5,223 firm-quarter observations on German-listed firms spanning 2010:Q1–2021:Q4. This study regresses the cost of debt financing on the geopolitical risk, accounting quality and other control variables. Information asymmetry is measured using the performance-matched Jones-model discretionary accrual and the stock bid-ask spread. It uses interaction terms to check if information asymmetry moderates the impact of geopolitical uncertainty on the cost of debts and control for the moderating role of business risk. For the sake of robustness check, it uses long-term cost of debt and bond spread as alternative dependent variables. In addition, this study executes instrumental variables regression and propension score matching to control for potential endogeneity problems.

Estimation results show that geopolitical uncertainty exerts a positive impact on the cost of debt. This impact is found to be more important on the cost of long-term debts. Information asymmetry is found to exacerbate the positive impact of geopolitical risk on the cost of debt. These results are robust to the change of the dependent variable and to the mitigation of potential endogeneity. At high levels of information asymmetry, this impact is more important for firms belonging to “Transportation”, “Automobiles and auto parts”, “Chemicals”, “Industrial and commercial services”, “Software and IT services” and “Industrial goods” business sectors.

Geopolitical uncertainty should be seriously considered when setting strategies for corporate financial management in Germany and similar economies that are directly exposed to geopolitical risks. Corporate managers should design a comprehensive set of corporate policies to improve their transparency and accountability during increasing uncertainty. Policymakers are required to implement innovative monetary and fiscal policies that take into consideration the heterogeneous impact of geopolitical uncertainty and information transparency in order to contain their incidence on German business sectors.

Despite its relevance to corporate financing conditions, little is known about the impact of geopolitical uncertainty on the cost of debt financing. To the best of the authors’ knowledge, there is still no empirical evidence on how information asymmetry between creditors and borrowing firms shapes the impact of geopolitical uncertainty on the cost of debt. This paper tries to fill this gap by interacting two measures of information asymmetry with geopolitical uncertainty. In contrast with previous studies, this study shows that the impact of geopolitical uncertainty on the cost of debt is non-linear and heterogeneous. The results show that the impact of geopolitical uncertainty does not exert the same impact on the cost of debt instruments with different maturities. This impact is found to be heterogeneous across business sectors and to depend on the level of information asymmetry.

The purpose of this paper is to compare nine different models to evaluate consumer credit risk, which are the following: Logistic Regression (LR), Naive Bayes (NB), Linear Discriminant Analysis (LDA), k-Nearest Neighbor (k-NN), Support Vector Machine (SVM), Classification and Regression Tree (CART), Artificial Neural Network (ANN), Random Forest (RF) and Gradient Boosting Decision Tree (GBDT) in Peer-to-Peer (P2P) Lending.

The author uses data from P2P Lending Club (LC) to assess the efficiency of a variety of classification models across different economic scenarios and to compare the ranking results of credit risk models in P2P lending through three families of evaluation metrics.

The results from this research indicate that the risk classification models in the 2013–2019 economic period show greater measurement efficiency than for the difficult 2007–2012 period. Besides, the results of ranking models for predicting default risk show that GBDT is the best model for most of the metrics or metric families included in the study. The findings of this study also support the results of Tsai et al. (2014) and Teplý and Polena (2019) that LR, ANN and LDA models classify loan applications quite stably and accurately, while CART, k-NN and NB show the worst performance when predicting borrower default risk on P2P loan data.

The main contributions of the research to the empirical literature review include: comparing nine prediction models of consumer loan application risk through statistical and machine learning algorithms evaluated by the performance measures according to three separate families of metrics (threshold, ranking and probabilistic metrics) that are consistent with the existing data characteristics of the LC lending platform through two periods of reviewing the current economic situation and platform development.

The monetary policy implementation and corporate investment are closely intertwined. The aim of modern monetary policy is to mitigate economic fluctuations and stabilise economic growth. One of the ways of influencing the real economy is influencing the level of investment by enterprises.

This chapter provides evidence on how monetary policy affected corporate investment in Poland between 1Q 2000 and 3Q 2022. We investigate the impact of Polish monetary policy on investment outlays in contexts of high uncertainty.

Using the correlation analysis and the regression model, we show the relation between the monetary policy and the investment outlays of Polish enterprises. We used the least squares method as the most popular in linear model estimation. The evaluation includes model fit, independent variable significance and random component, i.e. constancy of variance, autocorrelation, alignment with normal distribution, along with Fisher–Snedecor test and Breusch–Pagan test.

We find that Polish enterprises are responsive to changes in monetary policy. Hence, the corporate investment level is correlated with the effects of monetary policy (especially with the decision on the central bank's basic interest rate changes). We found evidence that QE policy has a positive impact on Polish investment outlays. The corporate investment in Poland is positively affected by respective monetary policies through Narodowy Bank Polski (NBP) reference rate, inflation, corporate loans, weighted average interest rate on corporate loans.

This study aims to investigate the efficiency of Indian commercial banks from 2002 to 2018 using the stochastic frontier analysis.

This study uses the parametric approach of the stochastic frontier to examine the technical efficiency of banks acknowledging exogenous shocks, omitted variables and measurement errors, filling a gap in the existing financial literature. The scope of this study was constrained to 71 scheduled commercial banks to make it manageable and productive with 1,036 observations.

The results show that the mean technical efficiency of new private banks remained constant at 92.7% during the study period because of technology diffusion in banking systems. The technical efficiency of the nationalized, old private and foreign banks has enhanced over the period because of the efficient utilization of various innovative information technology services such as mobile banking, cheque truncation system, magnetic ink character recognition. However, the foreign banks are still laggards with a mean technical efficiency of 81.7%. The empirical findings suggest that new private sector banks depict higher efficiency than nationalized, old private and foreign banks.

This study’s sample represents all categories of banks (public, private and foreign) including the banks that merged or consolidated during the period of study. To achieve the desired results, the authors incorporate the consolidated and merged banks in their data set. Further, the authors excluded all scheduled small finance banks and scheduled payment banks from their analysis, as these entities commenced operations post-2015. Additionally, the authors also excluded regional rural banks because of their distinct mandate aimed at servicing the rural populace and agricultural sector.

This study contributes to the literature on the performance of conventional banks in general and emerging markets, in particular, using the most recent data and covering a relatively long period using the stochastic frontier approach.

This paper aims to explore the dynamic relationship between ESG scores and REITS returns. The overarching goal is to provide a better understanding of how ESG considerations impact financial performance across different temporal contexts.

Using a sample of 175 European Equity REITs, this analysis combines numerical ESG scores with the Fama-French model, employing both random and fixed effects methods. It integrates individual REIT data and the HESGL (High ESG Scores Minus Low ESG Scores) factors to assess their impact on REIT returns.

The findings highlight divergent patterns between the numerical ESG score and the HESGL factor concerning REIT returns. While the numerical ESG score displays a negative impact in later periods, the HESGL factor demonstrates a positive effect during prosperous times but loses significance during crises.

This research contributes original insights by emphasizing the importance of temporal segmentation in understanding the nuanced and evolving nature of the relationship between ESG scores and REITs’ returns. The study provides a comprehensive analysis and highlights divergent outcomes that are essential for a better interpretation of ESG impacts on real estate investments.

The purpose of this study is to address a highly heterogeneous rift margin environment and exhibit considerable spatiotemporal hydro-climatic variations. In spite of limited, random and inaccurate data retrieved from rainfall gauging stations, the recent advancement of satellite rainfall estimate (SRE) has provided promising alternatives over such remote areas. The aim of this research is to take advantage of the technologies through performance evaluation of the SREs against ground-based-gauge rainfall data sets by incorporating its applicability in calibrating hydrological models.

Selected multi satellite-based rainfall estimates were primarily compared statistically with rain gauge observations using a point-to-pixel approach at different time scales (daily and seasonal). The continuous and categorical indices are used to evaluate the performance of SRE. The simple scaling time-variant bias correction method was further applied to remove the systematic error in satellite rainfall estimates before being used as input for a semi-distributed hydrologic engineering center's hydraulic modeling system (HEC-HMS). Runoff calibration and validation were conducted for consecutive periods ranging from 1999–2010 to 2011–2015, respectively.

The spatial patterns retrieved from climate hazards group infrared precipitation with stations (CHIRPS), multi-source weighted-ensemble precipitation (MSWEP) and tropical rainfall measuring mission (TRMM) rainfall estimates are more or less comparably underestimate the ground-based gauge observation at daily and seasonal scales. In comparison to the others, MSWEP has the best probability of detection followed by TRMM at all observation stations whereas CHIRPS performs the least in the study area. Accordingly, the relative calibration performance of the hydrological model (HEC-HMS) using ground-based gauge observation (Nash and Sutcliffe efficiency criteria [NSE] = 0.71; R² = 0.72) is better as compared to MSWEP (NSE = 0.69; R² = 0.7), TRMM (NSE = 0.67, R² = 0.68) and CHIRPS (NSE = 0.58 and R² = 0.62).

Calibration of hydrological model using the satellite rainfall estimate products have promising results. The results also suggest that products can be a potential alternative source of data sparse complex rift margin having heterogeneous characteristics for various water resource related applications in the study area.

This research is an original work that focuses on all three satellite rainfall estimates forced simulations displaying substantially improved performance after bias correction and recalibration.

Given the growing emphasis among scholars, practitioners and policymakers on financial sustainability, this study aims to explore the applicability of machine learning techniques in predicting the financial performance of microfinance institutions (MFIs).

This study gathered 9,059 firm-year observations spanning from 2003 to 2018 from the World Bank's Mix Market database. To predict the financial performance of MFIs, the authors applied a range of machine learning regression approaches to both training and testing data sets. These included linear regression, partial least squares, linear regression with stepwise selection, elastic net, random forest, quantile random forest, Bayesian ridge regression, K-Nearest Neighbors and support vector regression. All models were implemented using Python.

The findings revealed the random forest model as the most suitable choice, outperforming the other models considered. The effectiveness of the random forest model varied depending on specific scenarios, particularly the balance between training and testing data set proportions. More importantly, the results identified operational self-sufficiency as the most critical factor influencing the financial performance of MFIs.

This study leveraged machine learning on a well-defined data set to identify the factors predicting the financial performance of MFIs. These insights offer valuable guidance for MFIs aiming to predict their long-term financial sustainability. Investors and donors can also use these findings to make informed decisions when selecting their potential recipients. Furthermore, practitioners and policymakers can use these findings to identify potential financial performance vulnerabilities.

This study stands out by using a global data set to investigate the best model for predicting the financial performance of MFIs, a relatively scarce subject in the existing microfinance literature. Moreover, it uses advanced machine learning techniques to gain a deeper understanding of the factors affecting the financial performance of MFIs.

The study aims to examine the dividend omissions and dividend cuts behaviour of manufacturing and non-financial services firms to identify the determinants of dividend omissions and dividend cuts.

The study analyses the financial data of 3,546 firms from 2011 to 2020 (35,460 firm-year observations) using a dynamic random-effect probit panel regression model.

The results suggest that profitability, growth opportunity, leverage, liquidity, risk, extraordinary income, shareholding pattern and buyback are major determinants of dividend omissions. Similarly, dividend cut in the previous year, profitability, operating cash flow, risk and extraordinary income are major factors leading to dividend cuts.

Firms which omit the dividend are less likely to start paying dividend in subsequent years, whereas firms which cut the dividend may increase dividend in later years. Also, profitability decreases for a significant number of firms post dividend omission and cut. This indicates that dividend omission is a more prominent signal than a dividend cut for the financial health of a firm.

The determinants identified in the study enable analysts and portfolio managers to decide the propensity of dividend omission and cut even before actual announcements and can alleviate the significant loss in the portfolio. Also, managers and the board of directors would be able to monitor the firm’s financial performance to avoid the situation leading to dividend omissions and cuts.

The study strongly recommends that firms should voluntarily pay dividends to shareholders to encourage the healthy participation of retail shareholders in the equity market and create a long-term win–win situation for all stakeholders in society. If a large number of firms continue not to pay the dividend, the study appeals to the regulators to intervene to protect shareholders' interests for the greater good of society.

To the best of author’s knowledge, this is the first study to empirically identify the determinants of dividend omission and cut in the unique setting like India where dividend taxation had undergone a significant change.