Research On Stock Return Rate Based On Mining Financial Ratios Of China A Stock

Stock pricing is an enduring topic in the financial sector.There are currently several pricing models generally recognized by academia,such as capital asset pricing,Fama-French three-factor and five-factor,and Carhart four-factor pricing models.These pricing models can certainly play a role in explaining the stock return rate,but the stock return rate affects more than the factors contained in these models,so many scholars are committed to studying the influencing factors outside the model,trying to find out "Return rate anomalies",these "return rate anomalies" can provide directions for the improvement of pricing models,and "anomalies" related to financial indicators are an important part of them.At present,the financial indicators involved in the "Financial Indicators Anomalies" discovered by scholars at home and abroad are concentrated on common financial indicators.This has two shortcomings: First,there is the problem of "overperspective data",that is,a large number of scholars focus on Some common financial indicators,even if there is no specific relationship between these financial indicators and stock returns,but by limiting the research population,modifying the number of samples and changing the sample period,etc.,the logical relationship can finally be obtained;the second is the use of information Insufficient,a lot of accounting information is disclosed in the company's financial report.Common financial indicators only reflect a part of it.For the improvement of the pricing model,it is necessary to use as much information as possible to find a breakthrough.Therefore,in the face of the above two shortcomings,there is a need for a method of mining financial indicators that can avoid the problem of "over-perspective data" and use accounting information as much as possible.This article draws on the ideas of Xuemin(Sterling)Yan(2017),uses as much of the accounting information in the annual reports of China's listed companies as possible,avoids adding unnecessary qualifications as much as possible,and constructs more than seven thousand financial structures through simple function index.For each financial indicator,at the end of June each year,according to the size of the financial indicator at the end of the previous year,the sample stocks are divided into four stock combinations: low financial indicator combination(group L),group 2,group 3 and high financial indicator combination(Group H).Calculate the monthly weighted return rate of each stock combination from July to June of the following year based on the weight of the circulating market value,and so on,to obtain a total of 132 stocks in the four stock combinations from July 2008 to June 2019 The monthly return rate series(if year-onyear data is involved,only the 120-month return rate from July 2009 to June 2019).By paired sample t test,it was found that the monthly average returns of Group L and Group H with a total of 1,556 financial indicators were not equal at the 5% significance level.Subsequently,this paper brings the monthly returns of the four stock portfolios obtained by these 1556 financial indicators into the traditional pricing model during the sample period,so that both Group L and Group H can obtain excess positive at the 5% significance level Rate of return or negative rate of return,capital asset pricing model,Fama-French three-factor pricing model,Carhart four-factor pricing model,three Fama-French five-factor pricing models(2 * 3 grouping method,2 * 2 grouping,respectively)Method and the five factors calculated by the 2 * 2 * 2 * 2 grouping method)were screened out 6,6,46,15(2 * 3 grouping method),11(2 * 2 grouping method)and 4(2 * 2 * 2 * 2 grouping method)financial indicators.After research,it is found that most of the financial indicators selected by each pricing model are related to the "cash ratio".This paper selects five financial indicators related to "cash ratio" and cross-groups them with company market value,book market value ratio,profitability and investment level to construct an investment portfolio.Taking the market value of a company as an example,firstly,the sample stocks are divided into 4 groups according to the size of “cash ratio”: low cash ratio combination(group L),group 2,group 3 and high cash ratio combination(group H),followed by,The sample stocks are divided into 4 groups according to the market value of the company: small market value combination(group S),group 2,group 3 and large market value combination(group B),and finally 4 cash ratio combinations and 4 company size combinations Perform cross-fetching and intersection to form 16 investment portfolios,such as a small market value high cash ratio portfolio,a small market value low cash ratio portfolio,etc.By analogy,20 such 16 portfolios were formed.Calculate the monthly weighted rate of return of each portfolio with the market value of circulation as the weight,and finally bring the series of rate of return of each portfolio into the traditional pricing modelIt was finally found that the stock portfolio with a small total market value of the company and a high ratio of "end-of-period balance of cash and cash equivalents / total company market value" obtained the highest excess monthly rate of return in each pricing model due to 1)the high "cash to market value ratio" The overall valuation is lower.The “cash to market value ratio” portfolio has a lower overall valuation.2)Small-cap companies have good growth,but the financing constraints they face are more prominent than larger-cap companies.“Cash accounts for The “high market value ratio” has alleviated the financing constraints faced by small market capitalization companies,enabling them to invest in a timely manner when good investment opportunities arise,which in turn leads to improved performance,and improved performance will also bring the share price of these small market capitalization companies rise.