Journal
COMPUTATIONAL ECONOMICS
Volume 50, Issue 1, Pages 141-159Publisher
SPRINGER
DOI: 10.1007/s10614-016-9585-0
Keywords
Online portfolio selection; Online learning; Expert advice; Weak aggregating algorithm
Funding
- National Nature Science Foundation of China [71501049, 71301029]
- Humanities and Social Science Foundation of the Ministry of Education of China [11YJC630255, 13YJC630234]
- 125 planning for the development of Philosophy and Social Sciences in Guangzhou [15G29]
Ask authors/readers for more resources
The weak aggregating algorithm (WAA) developed from learning and prediction with expert advice makes decisions by considering all the experts' advice, and each expert's weight is updated according to his performance in previous periods. In this paper, we apply the WAA to the online portfolio selection problem. We first consider a simple case in which the expert advice is the strategy for investing in one stock; for this case, we obtain a portfolio selection strategy WAAS and prove that the WAAS can identify the best stock. We also discuss a more complicated case in which constant rebalanced portfolios are considered as expert advice, and obtain a corresponding portfolio selection strategy WAAC. The theoretical result shows that the cumulative gain that WAAC achieves is as large as that of the best constant rebalanced portfolio. Numerical analysis shows that the cumulative gains of our proposed strategies are as large as those of the best expert advice.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available