Research on the marginal effectiveness of Chinese stock markets'pricing factors:Application of double-selection LASSO model
In the present era of high-dimensional data,it is quite unlikely that the traditional methods for estimating pricing factors are capable of judging accurately the marginal effectiveness of pricing factors applicable to the Chinese stock markets.Hence we construct a double-selection LASSO model,instead of the traditional methods which estimate stock pricing factors mainly by estimating risk premium.And then by means of this double-selection LASSO model,we estimate stochastic discount factors loading,thereby being able to judge accurately the marginal effectiveness of stock pricing factors while processing high-dimensional data.Next,we gather together 85 pricing factors applicable to the Chinese stock markets,thus building up a high-dimensional pricing factor zoo.In addition,we identify 7 marginally effective pricing factors out of the 15 factors discovered after 2014.Our discovery proves consistent in various robustness tests.We further find in the follow-up analysis that the effectiveness of these 7 pricing factors proves consistent under the conditions of time-varying SDF.
NETL
NSTL
万方数据
