A股数据day级前复权数据下载与存储

Posted 2020-10-18 坴戋

存储为3个文件夹：大盘数据、股票数据与概括文件

数据获取来源：股票数据来源于R中的WindR；大盘数据来源于python中的tushare（wind中指数似乎只有中证板块）

设置数据文件夹，代码中只需修改更新截止日期与工作路径，后面创建文件都无需手动操作

文件夹中文件不能随意变动，标红代码部分需要自己修改

股票数据中变量分别为股票代码、股票简称、日期、时间、开高低收（乘10000）、交易量、交易额、转手率（基于自由流通股本）、自由流通股本

初次获取数据

1、大盘指数数据（399001, 399005, 399006, 399300，999999）

## 初始设置
os.chdir("C:/Users/Administrator.USER-20161208UW/Desktop/数据") #修改当前工作目录
start_date = "2009-01-01"
end_date = "2017-12-20"

new_date = end_date.replace("-", "")
## 创建新文件夹
os.mkdir("ts大盘数据")
os.mkdir("ts大盘数据/大盘数据"+new_date)

## 概括信息
stock_info = pd.DataFrame({"stock.code":["399001", "399005", "399006", "399300", "999999"] ,
"stock.name":["深证成指", "中小板指", "创业板指", "沪深300", "上证指数"],
"type":["SZ", "SZ", "SZ", "SZ", "SH"]})
stock_info.to_csv("ts大盘数据/大盘数据"+new_date+"/stock_info.csv", index=False)

# 获取大盘信息
dapanzhishu1 = ["399001", "399005", "399006", "399300", "000001"]
dapanzhishu2 = ["399001", "399005", "399006", "399300", "999999"]
wind_code = ["399001.SZ", "399005SZ", "399006SZ", "399300SZ", "999999SH"]
name = ["深证成指", "中小板指", "创业板指", "沪深300", "上证指数"]
for i in list(range(5)):

df = ts.get_h_data(dapanzhishu1[i], start=start_date, end=end_date, index=True, pause=4)
df.sort_index(inplace=True)
df[["open", "high", "close", "low"]] = df[["open", "high", "close", "low"]]*10000
df["date"] = df.index
df["date"] = df["date"].astype(str).apply(lambda x:x.replace(\'-\', \'\'))
df.columns = [\'open\',\'high\', \'close\', \'low\', \'volumw\', \'turover\', \'date\']
df["name"] = name[i]
df["wind_code"] = wind_code[i]
df["time"] = 151500000
df["turn"] = 0
df["free_turn"] = 0
df = df[["wind_code", "name", "date", "time", "open", "high", \'low\', \'close\', \'volumw\', \'turover\', "turn", "free_turn"]]
df.to_csv("ts大盘数据/大盘数据"+new_date+"/"+dapanzhishu2[i]+".csv", index=False)

2、A股股票数据与概括文件

## R
library(WindR)
library(xlsx)
library(data.table)
library(magrittr)
library(tcltk2)
setwd("C:/Users/~~~/Desktop/数据")

new.date <- "20171220"
start.date <- "20090101"

w.start()
end.date <- paste(substr(new.date, 1, 4), substr(new.date, 5, 6), 
                  substr(new.date, 7, 8), sep = "-")
start.date <- paste(substr(start.date, 1, 4), substr(start.date, 5, 6), 
                    substr(start.date, 7, 8), sep = "-")
## 摘取当日在市股票代码
stock.code.df <- w.wset(\'sectorconstituent\', 
                        paste0("date=", end.date, ";sectorid=a001010100000000"))
if(stock.code.df$ErrorCode == 0){
  stock.code.sh.sz <- stock.code.df$Data$wind_code
}else{
  print(paste0("获取数据出错，错误代码", stock.code.df$ErrorCode))
}
new.stock.code <- substr(stock.code.sh.sz, 1, 6)
# str(stock.code.sh.sz)

## 创建新的文件夹
dir.create("概括文件")
dir.create("股票数据")
dir.create(paste0("股票数据/股票数据", new.date))


## 补全stock.code的开始及截至日期及其他信息CSV文件
general.information <- data.frame(array(dim=c(length(stock.code.sh.sz), 5)))
colnames(general.information) <- c("stock.code", "stock.name", "type", "starttime", "endtime")


## 设置进度条
pb <- tkProgressBar("进度", "已完成 %", 0, 100) 

## 从WindR获取数据
## 未检查数据是否存在异常日期（大盘中没有的交易日）
w.start()
# i <- 1
for(i in 1:length(stock.code.sh.sz)){
  wind.data <- w.wsd(stock.code.sh.sz[i], "trade_code, sec_name, open, high, low, close, volume, 
                                           amt, free_turn,free_float_shares",
                     start.date, end.date, "unit=1;PriceAdj=F")
  if(wind.data$ErrorCode == 0){
    wind.df <- data.frame(array(dim=c(nrow(wind.data$Data), 12)))
    colnames(wind.df) <- c("wind_code", "name", "date", "time", "open", "high", "low", "close", 
                           "volumw", "turover", "free_turn", "free_float_shares")
    wind.df[, 1]    <- wind.data$Code
    wind.df[, 2]    <- wind.data$Data$SEC_NAME
    wind.df[, 3]    <- gsub("-", "", wind.data$Data$DATETIME)
    wind.df[, 4]    <- 151500000
    wind.df[, 5:8]  <- wind.data$Data[4:7] * 10000
    wind.df[, 9:12] <- wind.data$Data[8:11]
    wind.df <- wind.df[!(is.na(wind.df$open)), ]
    if(nrow(wind.df) == 0){
      print(c(i, stock.code[i]))
    }else{
      if(any(is.na(wind.df))){
        print(paste(stock.code.sh.sz[i], "数据出错（数据中仍有NA）"))
      }else{
        if(any(wind.df[, 5:8] == 0)){
          print(paste(stock.code.sh.sz[i], "数据出错（数据中开高低收存在0）"))
        }else{
          if(any(table(wind.df$date) > 1)){
            print(paste(stock.code.sh.sz[i], "数据出错（数据中存在日期相同）"))
          }else{
            if(any(wind.df$date != sort(wind.df$date))){
              print(paste(stock.code.sh.sz[i], "数据出错（数据中日期顺序不对）"))
            }else{
              general.information[i, 1] <- substr(stock.code.sh.sz[i], 1, 6)
              general.information[i, 2] <- wind.df$name[1]
              general.information[i, 3] <- substr(stock.code.sh.sz[i], 8, 9)
              general.information[i, 4] <- wind.df[1, 3]
              general.information[i, 5] <- wind.df[nrow(wind.df), 3]
              write.csv(wind.df, paste("股票数据/股票数据", new.date, "/", new.stock.code[i], ".csv", sep=""), 
                        row.names = FALSE)
            }
          }
        }
      }
    }
  }else{
    print(paste0(stock.code.sh.sz[i], "股票获取数据出错，错误代码：", wind.data$ErrorCode))
    break
  }
  info <- sprintf("已完成 %d%%", round(i*100/length(stock.code.sh.sz)))
  setTkProgressBar(pb, i*100/length(stock.code.sh.sz), sprintf("进度 (%s)", info), info)
}

## 关闭进度条
close(pb)

## 总概括文件中无NA时输出
if(all(!(is.na(general.information)))){
  write.xlsx(general.information, paste0("概括文件/概括文件", new.date, ".xlsx"), row.names = FALSE)
}else{
  print("总概况文件中存在NA，需查验")
}

　3、文件格式

（1）一级

（2）二级

 

 （3）三级