We describe an agent-based model of a market where traders exchange a risky asset whose returns can be partly predicted purchasing a costly signal. The decision to be informed (at a cost) or uninformed is taken by means of a simple learning by imitation mechanism that periodically occurs. The equilibrium is characterized describing the stationary distribution of the price and the fraction of the informed traders. We find that the number of agents who acquire the signal decreases with its cost and with agents’ risk aversion and, conversely, it increases with the signal-to-noise ratio and when learning is slow, as opposed to frequent. Moreover, price volatility appears to directly depend on the fraction of informed traders and, hence, some heteroskedasticity is observed when this fraction fluctuates.

Asymmetric information and learning by imitation in agent-based financial markets

Gerotto L.;Pellizzari P.;Tolotti M.
2019-01-01

Abstract

We describe an agent-based model of a market where traders exchange a risky asset whose returns can be partly predicted purchasing a costly signal. The decision to be informed (at a cost) or uninformed is taken by means of a simple learning by imitation mechanism that periodically occurs. The equilibrium is characterized describing the stationary distribution of the price and the fraction of the informed traders. We find that the number of agents who acquire the signal decreases with its cost and with agents’ risk aversion and, conversely, it increases with the signal-to-noise ratio and when learning is slow, as opposed to frequent. Moreover, price volatility appears to directly depend on the fraction of informed traders and, hence, some heteroskedasticity is observed when this fraction fluctuates.
2019
Communications in Computer and Information Science
File in questo prodotto:
File Dimensione Formato  
paams2019Tolotti_gerotto_pellizzari.pdf

accesso aperto

Descrizione: Intestazione volume e articolo
Tipologia: Documento in Post-print
Licenza: Accesso gratuito (solo visione)
Dimensione 549.64 kB
Formato Adobe PDF
549.64 kB Adobe PDF Visualizza/Apri

I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10278/3731417
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? ND
social impact