Project

Back to overview

New methods for asset pricing with frictions

English title New methods for asset pricing with frictions
Applicant Trojani Fabio
Number 197765
Funding scheme Project funding (Div. I-III)
Research institution Département des Sciences économiques Université de Geneve
Institution of higher education University of Geneva - GE
Main discipline Economics
Start/End 01.10.2020 - 30.09.2023
Approved amount 858'124.00
Show all

Keywords (5)

Asset Pricing; Stochastic Discount Factors; Gaussian Process Regression; Big Data; Market Frictions

Lay Summary (Italian)

Lead
Il progetto propone vari metodi per l'analisi teorica e empirica di mercati liberi da arbitraggio e con costi di transazione. I metodi sviluppati nel progetto danno luogo a strumenti utili per meglio comprendere le proprietà dei valori finanziari e il comportamento degli investitori in mercati finanziari con costi di transazione.
Lay summary
La prima parte del progetto studia approcci per testare modelli di valutazione e per selezionare fattori di rischio sistematico, in contesti diversi e in presenza di costi di transazione o errori di valutazione. La seconda parte considera problemi di ottimizzazione dinamica di portafoglio con un alto numero di strumenti di investimento e con costi di transazione, in equilibrio parziale e in equilibrio generale. Dal punto di vista empirico vengono sviluppate metodologie più realistiche per testare modelli di valutazione e per identificare fattori di rischio sistematico. Sul piano teorico, vengono proposte tecnologie di Machine Learning per risolvere problemi di portafoglio con costi di transazione per una classe più vasta di economie rispetto a quelle trattate finora in letteratura.

Direct link to Lay Summary Last update: 29.09.2020

Responsible applicant and co-applicants

Employees

Project partner

Abstract

This project proposes various new methods for the theoretical and empirical analysis of arbitrage-free asset markets in the presence of frictions and/or pricing errors. The first part of the project proposes new model-free technologies to (i) test asset pricing models in the presence of pricing errors, (ii) select asset pricing factors incorporating the no-arbitrage constraints implied by the Arbitrage Pricing Theory (APT) and (iii) detect global factors in international asset markets. The second part of the project focuses on dynamic economies with transaction costs. It proposes new technologies for solving optimal portfolio problems and for characterizing equilibrium asset prices in the presence of multiple traded assets and multivariate state dynamics. Empirically, we adopt linear pricing rules represented by stochastic discount factors that are compatible with general transaction cost structures. In this way, we introduce more realistic testing procedures for asset pricing models and factor detection approaches that are (i) economically motivated by assumptions regarding, e.g., market frictions or asymptotic no-arbitrage conditions in the APT and (ii) more suitable for the analysis of large asset markets. On the other side, we aim to develop technologies to solve partial equilibrium and general equilibrium asset pricing problems with transaction costs for a broader class of economic settings than previously achieved in the literature. By their nature, such technologies can give rise to powerful tools to improve the understanding of the properties of, e.g., asset prices, volume and trading behaviour in financial markets with frictions.
-