Research On Predictors Of Risk Premium

The stock market is an important part of financial market.The trend and fluctuation of stocks can reflect the political,economic and social conditions of a country and guide the country's macroeconomic regulation.With the development of the economy and the improvement of people's living standards,more and more people are beginning to use extra wealth for investment.Stocks are one of the most common investment methods.Therefore,stock market return forecasting is a very important financial issue.It is of great significance to the macroeconomic regulation and control of the country and the fact that investors buy and sell stocks and avoid risks to get the most benefit.However,the stock market's rate of return is affected by many factors.Among the many factors that cause the stock market to fluctuate,there are both internal market risk factors brought by the market participants' behavioral activities and external risk factors that have an impact on the market,such as politics,economy,company status,investor psychology and so on,so the stock price series is a complex system,it is difficult to predict it accurately.In the capital market,the research on the factors affecting asset prices and yields in the market began in the 1930 s and has always been a concern of scholars and investors at home and abroad.The risk premium,the difference between the expected rate of return and the risk-free rate,is extremely important for individual asset allocation and company investment decisions.In general,risk premiums are estimated by historical means,but historical means are often too smooth,resulting in fluctuations in certain periods that have not been applied to the corresponding forecasting work,and do not contain more available information.Therefore,historical mean is not a good estimate.The most popular predictor in this literature is the dividend–price ratio.Other economic variables that evince predictive ability include the earnings–price ratio,book-tomarket ratio,nominal interest rates,interest rate spreads,inflation,dividend-payout ratio,corporate issuing activity,consumption-wealth ratio,stock market volatility,labor income,aggregate output,output gap,expected business conditions,oil prices,lagged industry portfolio returns and accruals.This paper draws on the estimation method based on the maximum likelihood estimation proposed by Avdis and Wachter(2017),and applies the information contained in the forecasting factor net asset expansion ratio to the forecast of the US stock market return rate.The construction of the forecasting factor net asset expansion ratio takes into account factors such as stock market repurchase and issuance,while the predictive factor dividend ratio used by traditional academic research institutions only considers the payment of dividends and does not represent the entire cash flow of the company.After taking factors such as stock market repurchase and issuance into consideration,it can better measure the company's cash flow.The paper uses monthly and quarterly data about US stock market S&P 500 index from 1950 to 2017,the unconditional mean value obtained by the maximum likelihood estimation method has root mean square error that is significantly lower than it of the sample mean,and this result is statistically significant,which significantly increases the predictive power contained in the predictor itself.In addition to obtaining the most effective predictor for the US stock market,this paper also estimates the risk premium of Chinese stock market and makes recommendations for China's interest rate liberalization.