量化交易的单因子分析——聚宽平台

序言

对于公司基本面的单因子分析

检测方法

按照基本面的某一指标，按照大小排序，分成十个组别

代码

'''-------api调用聚宽-----'''
# 导入函数库
from jqdata import *
import datetime
import numpy as np
import pandas as pd


#模型选择
from sklearn.model_selection import train_test_split,cross_val_score

#数据处理
from sklearn.preprocessing import StandardScaler
# from sklearn.preprocessing import KBinsDiscretizer
#填充缺失值
# from sklearn.impute import SimpleImputer

#特征选择
from sklearn.feature_selection import SelectKBest,chi2,f_regression,SelectPercentile,RFE,SelectFromModel
from scipy.stats import pearsonr

# 模型
from sklearn.svm import SVR 
from sklearn.linear_model import LinearRegression,Ridge,Lasso,RandomizedLasso,LogisticRegression
from sklearn.cluster import KMeans
from sklearn.ensemble import RandomForestRegressor,RandomForestClassifier
from sklearn.tree import DecisionTreeClassifier

#调参
from sklearn.model_selection import GridSearchCV  
from sklearn.model_selection import learning_curve

#模型评估
from sklearn.metrics import silhouette_score,silhouette_samples,r2_score,roc_auc_score


# 初始化函数，设定基准等等
def initialize(context):
    # 设定沪深300作为基准
    g.security = list(get_all_securities(types=["stock"]).index)

    set_benchmark('000300.XSHG')
    # 开启动态复权模式(真实价格)
    set_option('use_real_price', True)
    # 输出内容到日志 log.info()
    # 过滤掉order系列API产生的比error级别低的log
    # log.set_level('order', 'error')
    
    ### 股票相关设定 ###
    # 股票类每笔交易时的手续费是：买入时佣金万分之三，卖出时佣金万分之三加千分之一印花税, 每笔交易佣金最低扣5块钱
    set_order_cost(OrderCost(close_tax=0.001, open_commission=0.0003, close_commission=0.0003, min_commission=5), type='stock')
    #设置滑点
    set_slippage(PriceRelatedSlippage(0.002))
      # 开盘前运行
    run_daily(before_market_open, time='before_open', reference_security='000300.XSHG')
      # 开盘时运行
    run_daily(market_open, time='open', reference_security='000300.XSHG')
      # 收盘后运行
    run_daily(after_market_close, time='after_close', reference_security='000300.XSHG')
     #检测因子
    g.factor=valuation.market_cap
    #分类数
    g.class_num=10
    g.day=0
    #时间窗口
    g.rate=22
    g.stock_class_list={
   }
    g.stock_class_close={
   }
    g.stock_class_return={
   }
    #收益率
    for i