The Construction Of Investor Attention Factor And Its Application Research

Investor attention has an important influence on the financial market,and it is a hot and meaningful topic in the field of behavioral finance.Many studies have shown that investor attention affects investors' behavior decision-making,thereby affecting asset pricing,leading to fluctuations in market returns,and a lot of attention risks.How to achieve effective pricing based on investor attention risk and improve market efficiency is an important and urgent problem as the investor attention risk is getting harder to ignore.In order to solve this problem,we need to build an asset pricing model considering investors' attention.Before that,of course,we need to explore whether investor attention can be constructed as an effective asset pricing factor.Therefore,it has great theoretical and practical significance for constructing an effective asset pricing factor,enhancing the market's asset pricing function,and improving market effectiveness.The objective of this paper is to construct the pricing factor of investor attention.By examining the relationship between investor attention and cross-sectional returns,this paper studies the asset pricing power of investor attention in financial markets from two perspectives of the change and the level,where the change in investor attention is a short-term change,while the level of investor attention is a continuous and cumulative long-term process.First of all,based on the background of investor attention risk in the real market,we put forward the research problems,summarize the relevant theories,give the research methods and technical routes,and point out the innovation of this paper.Secondly,we theoretically and empirically investigate the effects of the change in investor attention and the level of investor attention on stock returns.In the theoretical model,we study the relationship between the proportion of attentive investors,the propoetion of new attentive investors and the expected return in the two-stage economy with noise,respectively.Empirically,taking the number of users whose watch lists have the given stock provided by Eastmoney.com as the proxy variable of investor attention,we respective explore the correlation between the change in investor attention and the level of investor attention with stock returns.This paper confirms that investor attention has a significant impact on stock returns,and reveals that investor attention has a significant impact on asset pricing in the financial market.Thirdly,based on the cross-sectional level and time-series regression method,we comprehensively examine the asset pricing ability of the change in investor attention and the level of investor attention.From the cross-sectional level,we preliminarily explore the explanatory power of the change in investor attention and the level of investor attention,and confirm that the change in and the level of investor attention are the pricing factors that systematically affect stock returns.Based on the time series regression method,we make a more standardized exploration,that is,to construct the change in and the level of investor attention factors,and to investigate the explanatory power of the asset pricing model with attention factor for portfolio returns and market anomalies.It proves that the investor attention factor is an effective pricing factor,which improves the pricing efficiency of the financial market.Futhermore,we also explore the application of the above-mentioned two investor attention factors,and mainly examine the predictive ability of the change in and the level of investor attention factors on the overall market return.The last part is the conclusion,which summarizes the research content and shortcomings of this paper.The main conclusions are as below.1.The change in investor attention and the level of investor attention have significant impact on stock returns.First,according to theoretical derivation,we find that,with the increase of investors' attention allocation,the stock demand increases,which triggers an equilibrium price rise in the short term;on the other hand,when investor attention is regarded as the resource endowment of market information dissemination,more investors' attention assets will be possessed and the information transparency will be higher,which induce a low risk caused by information asymmetry and low risk premium compensation,so as to obtain lower expected return.Therefore,the change in attention has a positive relation with stock return,and the level of attention has a negative relation with stock return.Second,utilizing the watch number,we construct the investor attention return model,and empirically explore the above theoretical results,that is,in controlling other pricing factors,we regress the change in attention and the level of attention with individual stock returns.2.The change in investor attention and the level of investor attention are the pricing factors that significantly affect stock returns.First,the two variables of investor attention are cross grouped with size,book to market ratio and turnover,respectively.The results show that the explanatory power of the two attention to stock returns is not completely included by size,book to market ratio and turnover.Second,we use the data of the watch number to measure investor attention and select Shanghai and Shenzhen A shares with September 9,2013 to August 16,2016 as the research sample.Based on the Fama-Mac Beth two-step regression and rolling analysis,including four different data lengths and seven different calculation cycles of average return,we empirically show that the risk premiums of two attentions are significantly different from zero.Therefore,the change in attention and the level of attention are pricing factors.Investor attention can systemically cause individual investors' purchase behavior.The probability of an investor's purchase of a stock is related to his attention to the stock,so as to guide asset pricing.3.The asset pricing models augmented with two investor attention factors perform better than the base models both at capturing portfolio returns and market anomalies.Using the cumulative trading amount,which adjusted by the watch number,as a proxy variable for investor attention,and choosing the research sample of July,1996 to June,2018 in Shanghai and Shenzhen A-share stock market,we built a more general and consistent factors of the two attention.Then,we explore the pricing ability to portfolio returns and market anomalies of the attention-augmented model versus the base model with several evaluation criteria,which include the intercept of factor spanning test,maximum Sharp ratio,GRS statistics,absolute value of intercepts,absolute value of the intercepts divided by the absolute value of returns,and the estimated squared standard error of intercepts divided by absolute value of intercepts.The results show that the two attention factors are not redundant;the attention-augmented pricing models perform better than base models both at capturing the portfolio returns and market anomalies.These conclusions are also confirmed in building the size factor excludes the smallest 30% of firms,with which further supports the proposition that investor attention is an effective pricing factor.4.The factors of the change in attention and the level of attention have a significant predictive power for the aggregate stock market return.Using the data of two investor attention factors and the sample of August,1994 to December,2018 in Shanghai and Shenzhen A-share listed companies,we test its predictive power for aggregate stock market excess return in Chinese stock market.The out-of-sample predictability results show that the factors of two investor attention significantly predict the next month excess return;Compared with other predictors,two attention factors show a stronger predictive power.Our study not only enriches the relevant literature of investor attention,but also confirms the application value of investor attention factor in investment practice.