A minimax model of portfolio optimization using data mining to predict interval return rate
Meng Yuan, Junzo Watada
Abstract
Meng Yuan, Junzo Watada
Abstract
In 1950s, Markowitzs first proposed portfolio theory based on a mean-variance (MV) model to balance the risk and profit of decentralized investment. The two main inputs of MV are expected return rate and the variance of expected return rate. The expected return rate is an estimated value which is often decided by experts. Various uncertainty of stock price brings difficulties to predict return rate even for experts. MV model has its tendency to maximize the influence of errors in the input assumptions. Some scholars used fuzzy intervals to describe the return rate. However, there were still some variables decided by experts. This paper proposes a classification method to find the latent relationship between the interval return rate and the trading data of a stock and predict the interval of return rate without consulting any expert. Then this paper constructs the portfolio model based on minimax rule with interval numbers. The evaluation results show that the proposed method is reliable.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In 1950s, Markowitzs first proposed portfolio theory based on a mean-variance (MV) model to balance the risk and profit of decentralized investment. The two main inputs of MV are expected return rate and the variance of expected return rate. The expected return rate is an estimated value which is often decided by experts. Various uncertainty of stock price brings difficulties to predict return rate even for experts. MV model has its tendency to maximize the influence of errors in the input assumptions. Some scholars used fuzzy intervals to describe the return rate. However, there were still some variables decided by experts. This paper proposes a classification method to find the latent relationship between the interval return rate and the trading data of a stock and predict the interval of return rate without consulting any expert. Then this paper constructs the portfolio model based on minimax rule with interval numbers. The evaluation results show that the proposed method is reliable.
Key concepts: Rate of return on a portfolio, Portfolio, Rate of return, Minimax, Time-weighted return, Expected return, Holding period return, Econometrics