基于品牌资本因子的A 股投资策略研究
A -shares investment strategy research based on brand capital factor
谢振霖
Xie Zhenlin
分类号密级
U D C 编号
硕士学位论文
学位论文题目：基于品牌资本因子的A 股投资策略研究
姓名谢振霖
学号2021212319
学院中央财经大学粤港澳大湾区（黄埔）研究院
学位类别□学术硕士专业硕士□同等学力
学科专业金融硕士
指导教师张学勇
第二导师
提交论文日期：2023年5月16日
基于品牌资本因子的A 股投资策略研究
摘要
我国金融市场与国际上较为发达的金融市场相比，投资主体、市场机构、监管制
度都有所不同，市场上存在许多投资者寻找投资机会。其中资产定价一直是学术界与
业界研究的热点与重点，通过找寻合理的定价因子，发现市场上的定价异象，基于有
效的定价因子构建相应的投资策略，从而获得高额的收益。
近些年来，国外学者对于企业无形资本（Intangible Capital）的研究越来越多，无
形资本包括组织资本、品牌资本、研发资本等等区别于企业有形的资产。其中品牌资
本（Brand Capital）是企业重要的无形资本，可以表现消费者对于企业生产的商品和服
务的认知，拥有强势品牌的公司往往可以将自身产品和服务与竞争对手区分开来，由
此带来竞争优势，同时品牌资本存量会影响到公司的风险属性，从而影响其公司价值
以及股票收益。本文参考Belo（2014）和Hou 等人（2015）对于品牌资本的定义并结
合中国会计准则构建品牌资本因子（bca）。通过品牌资本因子构建投资组合，探究股
票收益与股票横截面收益是否存在相关关系，从而验证品牌资本因子在我国金融市场
是否有效。
为了研究品牌资本因子的有效性，本文根据品牌资本因子对A 股股票进行排序分
组，探究基于品牌资本因子构建的多空投资组合（购买bca 较高的股票并卖出bca 较低
的股票）的投资收益以及显著水平；接着进行多变量分析，探究基于品牌资本因子构
建的多空组合的超额收益能否被成熟的因子所解释；然后本文使用Fama-Macbeth 横截
面回归分析，检验在逐个增加控制变量的条件下，品牌资本因子对股票收益的预测作
用是否仍然有效；考虑到国有企业与非国有企业在品牌认知、财务约束方面存在差异，
因此本文将上市公司分为国企和非国企，探究品牌资本因子的预测效应是否存在企业
性质差异；另外受新冠疫情影响，疫情期间上市公司的正常营销活动、广告投入等无
法正常进行，本文剔除2020-2022年的疫情数据，同时增加了对于中国三因子，四因子
模型进行多变量分析，作为本文的稳健性检验；最后进行了策略回测，多头组合可以
获得15.33%的年化收益率，多空组合可以获得6.77%的年化收益率，并且也考虑了波
动率、夏普比率、最大回撤、月度胜率等方面，将多头组合、多空组合与沪深300指
数进行对比评价。
本文得到的主要结论如下：通过做多Higher 组（bca 较高的股票）并做空Lower
组（bca 较低的股票）能够获得高额的投资收益，具有统计学意义；在控制市场风险、
账面市值比风险、规模风险等一系列成熟因子后，基于bca 构建的多空投资组合依旧
存在超额收益；在Fama-Macbeth 横截面回归分析中发现，bca 因子值与股票未来收益
之间存在显著的正相关关系；bca 因子能够显著预测以民营企业为主的非国有企业上
市公司的股票收益，但对国有企业的预测效应不明显；与沪深300指数相比，基于bca
因子构建的多头组合和多空组合在样本期间均优于沪深300指数表现，本文的结果为
我国A 股市场投资者提供一定的借鉴与参考，具备现实意义。
关键词：品牌资本、横截面收益、资产定价、投资策略
Abstract
Compared with the more developed financial markets in the world,the main bodies of
investment,market institutions and supervision system are different.There are many investors
seeking investment opportunities in the market.Among them,asset pricing has always been
the hot spot and focus of academic and industry research.By searching for reasonable pricing
factors,we can find pricing anomalies in the market,and build corresponding investment
strategies based on effective pricing factors,so as to obtain high returns.
In recent years,there are more and more studies by foreign scholars on Intangible
Capital,including organizational capital,brand capital,R&D capital and so on,which are
different from tangible assets of enterprises.Brand Capital is an important intangible capital
of enterprises,which can reflect consumers'cognition of products and services produced by
enterprisespanies with strong brands can often distinguish their own products and
services from competitors,thus bringing competitive advantages.Meanwhile,brand capital
stock will affect the risk attributes of companies.That affects the value of the company and its
stock returns.This paper refers to the definition of brand capital by Belo (2014)and Hou et al.
(2015)and constructs brand capital factor (bca)in combination with Chinese accounting
standards.Through the construction of investment portfolio by brand capital factor,we
explore whether there is a correlation between stock return and stock cross section return,so
as to verify whether brand capital factor is effective in our financial market.
In order to study the effectiveness of the brand capital factor,this paper sorted and
grouped A-shares according to the brand capital factor,and explored the investment return
and significance level of the long-short portfolio constructed based on the brand capital factor
(buying stocks with higher bca and selling stocks with lower bca).Then,multivariate analysis
is carried out to explore whether the excess return of the long-short portfolio based on the
brand capital factor can be explained by mature factors.Then,this paper uses Fama-Macbeth
cross sectional regression analysis to test whether brand capital factor is still effective in
predicting stock returns under the condition of increasing control variables one by one.
Considering the differences between state-owned enterprises and non-state-owned enterprises
in brand cognition and financial constraints,this paper divides listed companies into state-
owned enterprises and non-state-owned enterprises to explore whether there are differences in