Study On Stock Selection Strategies Considering Digital Transformation Factors

The digital era has begun,and the development of the digital economy is an important strategic direction and a new engine of economic growth.According to the latest data from the National Bureau of Statistics,the scale of China’s digital economy reached 45.5 trillion yuan in 2021,accounting for 39.8% of the national GDP.As major participants in economic activity,enterprises have set off a wave of digital transformation.Nearly ninety percent of listed companies in China have carried out transformation activities,and nearly forty percent of enterprises have regarded it as their strategic core business,and digital transformation has become an important task for the success of the digital economy industrial revolution.Digital transformation aims to increase enterprise value and sustainability,which aligns with the focus of value investing.Therefore,through theoretical analysis,this thesis summarizes the impact of this factor on enterprise value,starts from the factors of enterprise digital transformation,explores the factors that can effectively explain the stock market,and explores its further connection with the capital market.The empirical part is first screened by factors,which are divided into two parts:selecting factors of digital transformation and constructing mainstream factor databases,digital transformation candidate factors are portrayed by three mainstream methods through text mining method,and the digital DTI factors in this thesis are constructed after validity verification,and the mainstream factors are summarized and included by referring to the major factor library.Then backtesting the strategy,The backtest time is2014.1-2021.12,a long strategy based on DTI factor is constructed,and the combination of equal weighted and tilted weighted positions is compared,After comparing the fitting effects of Ada Boost,XGBoost and Light GBM,XGBoost was selected to construct a factor integration strategy based on DTI,and the stock selection backtest results of the factor pool and the mainstream factor database integrating DTI factor are compared with the CSI 300 and active stock funds as benchmark indicators;finally the strategy is optimized,and based on the strategy performance analysis,the RSRS index and macro signal optimization results of the stop loss module RSRS index and macro signal are added respectively.The main conclusions of this thesis are: first,digital transformation factors have the ability to positively explain stock returns,and DTI factors constructed with digital intangible assets have stock selection capabilities;Second,digital transformation factors can obtain excess returns in volatile markets and bull markets,which are the most prominent and have certain resilience in bear markets,but the addition of this factor also brings increased risks.The XGBoost stock selection strategy after integrating DTI factors has a compound annualized return of 34.01% and a Sharpe ratio of 1.07,and the comprehensive indicators and yield indicators are better than the stock selection strategies constructed by the mainstream factor pool,and the indicators are better than the active stock selection funds.The empirical results show that this factor has an optimal effect on stock selection strategies,expands the traditional factor library,and its characteristics can provide reference for market participants and have reference significance.