Macroeconomic Factor Study Of APT Model Based On China's Stock Market

When individuals are in the face of investment options,they have to value expected return and assess risk premium of a variety of assets.And this is also the driving force behind the modern investment theories.Since Markowitz proposed portfolio theory,there extends out two categories around the identified risk factors.One is balanced pricing,which deepen the Markowitz's research;the other is arbitrage pricing,such as Ross's Arbitrage Pricing Theory(APT).In the guidance of Ross's APT,there developed a large number of multi-factor arbitrage pricing models.Classical ones divided into two fields:academic and business.The first field includes Fama-French's three-factor model and BIRR model;the latter involve RAM model created by Salomon Brothers and BARRA model of Morgan International Capital Corporation.All these models have been proved of practical application.This paper is triggered by the former points.Based on the research of factors,this paper studied macroeconomic factors.And the reasons can be attributed into three aspects:review of the previous literature,good effect of business models and the special circumstances of China's market.Meanwhile,in order to focus on important points,this paper chose macroeconomic factors as research emphasis.Taking into account that there always exist self-related phenomena among macroeconomic factors,this paper used Factor Analysis to eliminate self-relation and extract common factors.Firstly Chapter II introduced Arbitrage Pricing Theory and then took BIRR,RAM and BARRA as a reference,described the macroeconomic factors of these three models in details.As a result it classified five factors as follows:interest rate,exchange rate,inflation,economic growth and capital market factors.Based on the characteristics of China's stock market,this paper transferred these five factors into 15 Chinese macroeconomic factors:5-year deposit and loan spread,7-day interbank lending rate(monthly average),exchange rate,CPI,GDP,generating capacity,industrial added value,retail sales of total social customer goods,net export,fixed asset investment growth,new loan growth,industrial added value growth rate,Shanghai&Shenzhen 300 index,M1.Taking into account that macroeconomic factors may exist co linearity,Chapter ? extracted 5 common factors from 15 ones by the use of factor analysis.And these 5 common factors stand for the following meanings:economic development of overall,inflation level,development of the overall commercial activity,economic expansion and expectation for the future economy.So far this paper established a multi-factor arbitrage pricing model basing on the 5 common factors.One of the important applications of APT is to forecast portfolio returns and make investment decisions.Therefore in Chapter IV,this paper chose Shanghai and Shenzhen 300 index sample stocks to do empirical research and built China's stock market multi-factor arbitrage model.In the process of building a portfolio,this paper considered two ways:field and company size and then constructed two regression models.By the comparison of results,field model had better efficiency than company size one,but only minor differences had been showed.Finally,this paper carried two regression models for error analysis,that is to say use this model to predict return and compare it with actual income.Results showed that 22%and 21%error exited.This means all these two models can reflect the return of portfolios.So far this paper finished the task of establishing macroeconomic multi-factor APT model based on China's stock market.And it also gave description and demonstration of application and efficiency of this model.The last but not the least,Chapter V provided the direction of further study and methods to improve model.Most important one is to establish field macroeconomic factors.And it will receive more adequate results in this way.