Published on: 4th May 2025
Author: Gianni Parisi & Andrea Landini
[abstract]
Published on: 15 January 2025
Author: Andrea Landini
This study introduces a neural network framework for estimating parameters in jump-diffusion models of asset prices. Using OHLC data, the model refines key parameters (drift, volatility, jump intensity, and jump size) by minimizing the difference between simulated and actual prices. This approach effectively captures both continuous market trends and sudden jumps, providing a robust tool for modeling complex financial dynamics.
Published on: 21 January 2025
This study explores a dual-neural-network system for ETF analysis, integrating top holdings scoring and fund-level selection. Leveraging financial metrics such as EPS, EBITDA, and Net Income, the framework ensures comparability across ETFs and optimizes portfolio selection. The results demonstrate robust performance in aligning ETF choices with investment goals.
Published on: 10 November 2024
Author: Gianni Parisi, Andrea Landini
This paper examines carry trade strategies, focusing on market reactions to interest rate changes and currency interventions. It explores how carry trades leverage interest rate differentials, with investors borrowing in low-yield currencies to invest in higher-return assets. The study reviews types of carry trades, such as currency and asset-based, and discusses risks like currency, interest rate, and liquidity. The Dollar-Yen carry trade, affected by Bank of Japan interventions, illustrates the strategy's volatility.
Not Yet Published
This study investigates modeling gold price dynamics using Stochastic Models and Multivariate Markov-switching GARCH (MSGARCH), which capture volatility, regime shifts, and asset dependencies. MSGARCH identifies volatility phases, while stochastic models address long-term trends and sudden shifts.
Interest rate swaps (IRS) are vital tools for managing interest rate exposure, offering flexibility in cash flow and hedging strategies. This analysis employs stochastic volatility models and regime-switching GARCH to capture dynamic fluctuations, mean reversion, and volatility clustering in IRS markets.