Research on the Risk of Individual Stock Investment -- Take the Brokerage Sector as an Example

Zheng Ma, Hongmei Zhang, Ziyan Wang2016 · DOI

Abstract

As the stock market continues to evolve, individual investors have become a primary force in securities investment. However, these investors often face significant challenges in identifying valuable information amidst a sea of data, leading to suboptimal investment choices and low risk tolerance during market fluctuations. This study evaluates the investment risk of 15 brokerage stocks traded on the Shanghai Stock Exchange using SPSS-based factor analysis and cluster analysis. By ranking and grouping these stocks, the paper provides a quantitative framework to help individual investors mitigate risk and identify high-value opportunities.

Introduction

In the Chinese stock market, individual investors account for approximately 80% to 90% of trading volume. Despite their dominance in numbers, they frequently lack the professional tools required to process complex financial reports and daily market volatility. This "information asymmetry" often results in poor decision-making during periods of high volatility, such as the market corrections of 2007 and 2015. This research focuses on the brokerage sector due to its visible performance metrics and straightforward business models, serving as a test case for simplified, data-driven stock selection.

Methodology

The study utilizes a two-pronged statistical approach to evaluate 15 brokerage stocks based on financial data from early 2016.

1. Systematic Cluster Analysis

This method groups the sample stocks into four distinct categories based on similarities across seven core financial metrics (ROE, EPS, Asset-Liability Ratio, Total Assets, Operating Income, Net Profit, and Operating Profit). This allows for the identification of peer groups within the sector.

2. Factor Analysis

Using Principal Component Analysis (PCA), the researchers reduced the seven financial indicators into two primary dimensions:

  • Factor 1 (Profitability Factor): Captures the scale of revenue and profit generation.
  • Factor 2 (Yield Factor): Reflects the efficiency of the company in generating returns for shareholders.

To ensure statistical validity, the dataset was subjected to the Kaiser-Meyer-Olkin (KMO) test and Bartlett’s test of sphericity. The KMO score of 0.679 and a significance level of p < 0.000 confirmed that the data was well-suited for factor analysis.

Analysis of Variance

A critical finding of the factor analysis was the efficiency of dimensionality reduction. The study found that the first two components alone could explain the vast majority of the financial variance among the firms.

Total Variance Explained by Principal Components

As shown in the variance table, Factor 1 (Profitability) accounts for 56.23% of the variance, while Factor 2 (Yield) accounts for 37.10%. Together, these two factors explain 93.3% of the total variance, suggesting that investors can effectively evaluate stock quality by focusing on these two dimensions rather than tracking dozens of disparate metrics.

Results and Stock Rankings

By weighting the two factors based on their variance contribution rates, the researchers calculated a comprehensive score for each stock. This allowed for a definitive ranking of the 15 brokerage firms in terms of investment value and risk profile.

Comprehensive Factor Scores and Rankings

Key Findings:

  • Top Tier (Low Risk): CITIC Securities (4.71), Haitong Securities (4.53), and Guotai Junan (4.03) emerged as the clear leaders. These firms represent the "Cluster 3" group identified in the cluster analysis—high-quality, large-scale operations with stable returns.
  • Mid Tier: Firms like China Merchants Securities and Huatai Securities showed strong potential but require more strategic entry points.
  • High Risk: Anxin Trust (now Minmetals Capital) received the lowest score (1.02), indicating significantly higher risk for individual investors.

Conclusion

This research demonstrates that multivariate statistical techniques can effectively simplify the stock selection process for individual investors. By reducing complex financial data into "Profitability" and "Yield" factors, investors can identify sector leaders that offer a better balance of return and risk. While the study is limited by its sample size and static timeframe, the framework provides a scalable model for quantitative risk assessment across other market sectors.

Summary updated

Abstract

Today, Invest in the stock market has increasingly perfect continuous development, individual investors has gradually become the main force of stock investment, but due to various constraints, individual investors for face of massive investment information, they cannot identify the valuable investment intelligence information, choice of the value of the investment way are different degrees of defects and individual investors in the face of abnormal fluctuations in the stock market's risk bearing capacity is low. In order to identify individual investors in securities investment risk, this paper selects Shanghai Stock Exchange traded 15 brokerage stocks as a sample, using SPSS factor analysis and cluster analysis methods sample stock ranking and grouping, so as to identify the stock risk and for individual investors in securities investment provide relevant reference.
BibTeX
@misc{ma2016research,
  author = {Zheng Ma and Hongmei Zhang and Ziyan Wang},
  title = {{Research on the Risk of Individual Stock Investment -- Take the Brokerage Sector as an Example}},
  year = {2016},
  doi = {10.2991/rac-16.2016.83},
}
Figure