Seminar științific Monedă, Finanțe și Bănci – Arhiva 2026/2027

Dividend Policy with Interpretable Machine Learning

  • Marți, 10 noiembrie 2026 – ora 17:30 – sala 3607 (etaj 1, clădirea Moxa)

Codruț Ivașcu

Abstract:
Understanding why firms pay dividends, and how much, remains a central open question in corporate finance. Yet empirical research continues to rely largely on linear models, despite their limited predictive performance and the infrequent validation on out of sample. This paper proposes an interpretable machine learning framework that combines the predictive accuracy of gradient boosting with economically meaningful determinants characterized by direction, magnitude, and statistical significance. Using a large dataset we model four payout measures with XGBoost, evaluate them out of sample on an expanding window design, and interpret them with SHAP (SHapley Additive exPlanations) values. Refitting the whole pipeline on repeated subsamples of firms attaches statistical significance to each overall importance score and of the determinant’s impact at each quantile of its distribution. The models roughly double the predictive power of ordinary least squares and Tobit on held out years. Whether a firm pays and how much it pays turn out to be governed by different forces: firm scale and the country of incorporation drive the decision, while profitability and growth drive the amount. The country effect proves largely institutional, the leading fundamentals are non-linear, with effects concentrated in parts of their range, and past payout dominates the dynamic model. The results are robust to sample, specification, and the choice of learning algorithm.

Zoom:
https://ase.zoom.us/j/88975840175?pwd=Q0ZnbHl0TjhqcEtzVnhMUWRmdWdPQT09

Activitate de socializare: Zăganu Bistro

Beyond Traditional Financial Literacy: Digital Financial Literacy and Financial Well-Being

  • Marți, 27 octombrie 2026 – ora 17:30 – sala 3607 (etaj 1, clădirea Moxa)

Miruna Pochea

Abstract:
Digitalization is transforming the way individuals access and use financial services, raising the question of whether digital financial literacy contributes to financial well-being beyond traditional financial literacy. Using individual-level data from the 2023 OECD/INFE International Survey of Adult Financial Literacy for 24 countries, this study examines the relationships between traditional financial literacy, digital financial literacy, and financial well-being. The findings show a consistent positive relationship between traditional financial literacy and financial well-being. Digital financial literacy is also positively related to financial well-being after controlling for traditional financial literacy, although the relationship is substantially weaker. The two dimensions of financial literacy display distinct patterns across financial well-being: traditional financial literacy is more strongly linked to objective financial well-being, whereas digital financial literacy shows a closer relationship with subjective financial well-being. Moreover, the positive link between digital financial literacy and financial well-being is concentrated among individuals with higher levels of financial well-being and is not evident at the lower end of the distribution.

Zoom:
https://ase.zoom.us/j/88975840175?pwd=Q0ZnbHl0TjhqcEtzVnhMUWRmdWdPQT09

Activitate de socializare: Ever After

Macroprudential policy and French asset market responses: evidence from an ARDL Bounds Testing approach

  • Marți, 20 octombrie 2026 – ora 17:30 – sala 3607 (etaj 1, clădirea Moxa)

Andreea Stoian

Abstract:
This paper aims to investigate the effects of macroprudential policy measures on asset price dynamics, focusing on the banking sector stock market and residential house prices in France. Using quarterly data for the period 2000Q3 to 2024Q3, the anlaysis employs an Autoregressive Distributed Lag Bounds Testing framework to examine both the short and long-run relationships between macroprudential policy and the two markets. The empirical results suggest that the effects of macroprudential policy are not uniform across markets. While tighter macroprudential measures tend to reduce bank stock prices in the short run, the relationship with house price dynamics appears more complex.

Zoom:
https://ase.zoom.us/j/88975840175?pwd=Q0ZnbHl0TjhqcEtzVnhMUWRmdWdPQT09

Activitate de socializare: J’ai Bistrot

Cross-border banking, intragroup exposures and risk-taking

  • Marți, 13 octombrie 2026 – ora 17:30 – sala 3607 (etaj 1, clădirea Moxa)

Răzvan Vlahu

Abstract:
Regulatory limits on intragroup exposures constrain capital allocation within multi-national banking groups. We develop a theoretical model of cross-border banking that captures internal capital markets under supranational supervision and borrowing constraints. Our analysis shows that relaxing intragroup exposure limits can amplify risk-taking by enabling parent banks to draw on affiliate resources and reallocate risk toward the home market, particularly when foreign affiliates are large, well capitalized, and subject to weaker liquidity requirements. We characterize the conditions under which this channel operates and discuss its implications for financial stability. Our findings inform the debate on multinational banking groups by showing how risks can emerge within these organizations and how regulatory tools can mitigate them.

Zoom:
https://ase.zoom.us/j/88975840175?pwd=Q0ZnbHl0TjhqcEtzVnhMUWRmdWdPQT09

Activitate de socializare: Hop Hooligans Taproom

We are using cookies to give you the best experience. You can find out more about which cookies we are using or switch them off in privacy settings.
AcceptPrivacy Settings

GDPR

  • Cookie Consent

Cookie Consent

Anumite date pot fi procesate