Research On The Optimization Of Investment Portfolio Based On Improved Ant Colony Algorithm

Portfolio investment is aimed at scattering the risk and gaining income and allocating investment when the fund of investors is limited, to make the constructed portfolio achieve a certain income with minimum risk, or a maximum benefit with a certain risk. Due to the financial needs, investors invest more and more mone y in the stock market, but the stock investment is a high risk activity. Constructing portfolio rather than “putting all the eggs in one basket” is a way to effectively scatter the risk and obtain higher benefits.The ant colony algorithm is the intelligent optimization algorithm which simulates biological ants’ foraging behavior. Artificial ant colony algorithm has successfully solved a series of difficulties in our life such as quadratic assignment and traveling salesman problem and achieved better results. Ant colony algorithm has wide application and less restriction in solving problems. In recent years, some scholars have already used intelligent optimization algorithm to solve the problem of portfolio, but the study and application of ant colony algorithm in this area is comparatively less. Therefore, the application of improved ant colony algorithm to solving the problem of portfolio is of great practical significance.Based on the improvement of Markowitz’s classic portfolio model, this thesis adds the constraints of the smallest unit of stock trading, the in-process transaction cost and the maximum transaction volume to the model to make it of more practically significant. Portfolio is a way of finding the extremum of multi-objective function. To comb ine it with the ant colony algorithm, the thesis borrows from the Markowitz’s classic portfolio model the process of finding the solution and changes the multi-objective problem into one of finding the maximum value of a single-objective function by way of combined income minus portfolio variance. Starting with the basic ant colony algorithm, this thesis improves its information element and heuristic function and combines it with such indexes as earnings ratio, variance and volume which affect the prices of stocks so as to more efficiently optimize the portfolio model.Currently, the development of internet finance is bring about radical changes to our country’s finance market and financial products in internet finance have become one of people’s major choices. Therefore, this thesis takes Yu E Bao, a representative product of internet finance into the investment objective in the construction of portfolio model. In the empirical analysis, the paper selects five stocks listed in the internet and Yu E Bao as research subjects to obtain two portfolios at the maximum and minimum points of the objective function respectively. The thesis then analyzes the two resulting portfolios respectively in terms of income, risk, cost, and the suited group respectively. The analysis shows that the results obtained from the models are in line with normal stock market fluctuations as well as people’s investment habits.There are three points of innovation in this thesis. One is establishing a more practical portfolio model on the basis of summarizing others’ achievements; the second is taking Yu E Bao as an investment target to synchronize with the pace of financial market; the third is using the ant colony to simulate the behaviors of security investment in order to perfect the combination of the algorithm and the model and improve the optimization effect. The thesis then runs an analysis of the results obtained from the two portfolio models in terms of economics, which shows the results of the models are consistent with the actual results.