An adaptively managed dynamic portfolio selection model using a time-varying investment target according to the market forecast

Jongbin Jung, Seongmoon Kim

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

In this paper, we propose an adaptive investment strategy (AIS) based on a dynamic portfolio selection model (DPSM) that uses a time-varying investment target according to the market forecast. The DPSM allows for flexible investments, setting relatively aggressive investment targets when market growth is expected and relatively conservative targets when the market is expected to be less attractive. The model further allows investments to be liquidated into risk-free assets when the market forecast is pessimistic. By dynamically determining the investment target, the DPSM allows construction of portfolios that are more responsive to market changes, while eliminating the possibility of the model becoming infeasible under certain market conditions. When the proposed DPSM is implemented in real-life investment scenarios using the AIS, the portfolio is rebalanced according to a predefined rebalancing cycle and the model's input parameters are estimated on each rebalancing date using an exponentially weighted moving average (EWMA) estimator. To evaluate the performance of the proposed approach, a 7-year investment experiment was conducted using historical stock returns data from 10 different stock markets around the world. Performance was assessed and compared using diverse measures. Superior performance was achieved using the AIS proposed herein compared with various benchmark approaches for all performance measures. In addition, we identified a converse relationship between the average trading volume of a market and the value of the weighting parameter prescribed to the EWMA estimator, which maximizes cumulative returns in each market.

Original languageEnglish
Pages (from-to)1115-1131
Number of pages17
JournalJournal of the Operational Research Society
Volume66
Issue number7
DOIs
Publication statusPublished - 2015 Jul 16

All Science Journal Classification (ASJC) codes

  • Management Information Systems
  • Strategy and Management
  • Management Science and Operations Research
  • Marketing

Fingerprint Dive into the research topics of 'An adaptively managed dynamic portfolio selection model using a time-varying investment target according to the market forecast'. Together they form a unique fingerprint.

  • Cite this