Home bias in portfolio choices: social learning among partially informed agents

Wen Lin Wu, Yin Feng Gau

Research output: Contribution to journalArticlepeer-review

5 Scopus citations


Using an ‘incomplete information’ model, we explore the role of social learning in the global portfolio choices of stock market investors. When partially informed followers attempt to estimate true domestic (home) mean returns, they likely acquire private domestic signals from partially informed leaders. However, the calibration results indicate the existence of home bias when partially informed agents have poor quality information. Partially informed agents are prone to a learning bias; they overreact to new domestic information due to overconfidence in their domestic private signals, but they demonstrate a conservative response to new information in foreign markets. Links between the private signals of partially informed agents may lead to correlated foreign investment strategies among such agents through social learning. We suggest the acquisition of private signals, along with the dissemination of information, affect international portfolio decision rules and are determinants of the phenomenon of home bias.

Original languageEnglish
Pages (from-to)527-556
Number of pages30
JournalReview of Quantitative Finance and Accounting
Issue number2
StatePublished - 1 Feb 2017


  • Home bias
  • International asset allocation
  • Observational learning
  • Prior beliefs
  • Social learning


Dive into the research topics of 'Home bias in portfolio choices: social learning among partially informed agents'. Together they form a unique fingerprint.

Cite this