Efficiency Analysis of Machine Learning Intelligent Investment Based on K-Means Algorithm

Liang Li,Xuetao Li,Jia Wang

doi:10.1109/access.2020.3011366

Abstract

With the rapid development of technologies such as big data, intelligent data analysis and cloud computing, the application of Internet financial technology has become more and more extensive, and with the advent of the era of large asset management in the domestic wealth management industry, in order to improve the efficiency of financial services, traditional finance is needed. The products and services provided by the industry have been innovated, resulting in smart investment. Compared with traditional investment, smart investment as a new business model has the advantages of low threshold, low cost and high efficiency. However, as far as its nature is concerned, smart investment must first play the role of an investment adviser. Therefore, for enterprises or individuals who invest, the investment efficiency of smart investment is the most important. At present, the research on the efficiency analysis of smart investment, due to the improper selection of algorithm models or the lack of deep data mining, leads to the analysis of the investment efficiency of smart investment products is inconsistent with or even deviated from the actual situation. In view of these problems, this paper selects China Merchants Bank's Capricorn Intelligence as the research object, and analyzes the investment efficiency of smart investment based on K-means cluster analysis and data mining technology. The results show that Capricorn has a certain randomness in the selection process of the fund, and chooses to reduce the rate of return in order to control the risk. The investment portfolio formulated for the customer has obvious timing. The results show that the machine learning based on K-means algorithm makes a concrete analysis of the investment efficiency of Capricorn Smart Investment, this method can also be used for the efficiency analysis of other smart investment products.

Highlights

In the field of wealth management, traditional investment consultants need to complete a series of work such as investor risk preference analysis and asset allocation in a large number of manual ways, which invisibly raises the threshold of financial management services
Comparing the clustering results with the annualized rate of return results, it is found that the risk level of the 530008 fund is 4,110018, and the risk level of the fund is 4,233,013
The current development of smart investment is still in the exploratory stage, and it is facing the test of supervision, technology, market and so on

Summary

Introduction

In the field of wealth management, traditional investment consultants need to complete a series of work such as investor risk preference analysis and asset allocation in a large number of manual ways, which invisibly raises the threshold of financial management services. The management rate of traditional investment is about 1%, plus the operating expenses of ETF products and a series of other expenses, and the management fee rate of intelligent investment robots can be reduced to about 0.5%. This greatly reduces the cost of service for wealth management.

Methods

Results

Conclusion