将数据放入全球环境

时间:2018-07-16 20:13:16

标签: r

我以前有以下代码,效果很好,它使用quantmod包从Google下载财务信息。但是此功能此后已停止工作。值得庆幸的是,有人已经从Yahoo here获取数据。

library(tidyquant)
#symbols <- c("HOG", "GOOG", "GE")
    stockinfo <- tq_index("SP500")
    symbols <- stockinfo$symbol

    library(quantmod)
    symbols.f <- sapply(symbols, function(x) { paste0(x, ".f") })
    symbols <- sub("\\.f","", symbols.f)

    tickers <- new.env()
    lapply(symbols, getFinancials, env=tickers)
    BS <- data.frame(lapply(tickers, function(x) {viewFinancials(x, type= 'BS', period = 'A')}))
    IS <- data.frame(lapply(tickers, function(x) {viewFinancials(x, type= 'IS', period = 'A')}))
    CF <- data.frame(lapply(tickers, function(x) {viewFinancials(x, type= 'CF', period = 'A')}))

我正在应用以下代码,其中包含从yahoo下载财务信息的功能;

#########################################################
#Set the function to scrape Yahoo finance

getFin <- function(stock){
  if ("rvest" %in% installed.packages()) {
    library(rvest)
  }else{
    install.packages("rvest")
    library(rvest)
  }
  for (i in 1:length(stock)) {
    tryCatch(
      {
        url <- "https://finance.yahoo.com/quote/"
        url <- paste0(url,stock[i],"/financials?p=",stock[i])
        wahis.session <- html_session(url)                                
        p <-    wahis.session %>%
          html_nodes(xpath = '//*[@id="Col1-1-Financials-Proxy"]/section/div[3]/table')%>%
          html_table(fill = TRUE)
        IS <- p[[1]]
        colnames(IS) <- paste(IS[1,])
        IS <- IS[-c(1,5,12,20,25),]
        names_row <- paste(IS[,1])
        IS <- IS[,-1]
        IS <- apply(IS,2,function(x){gsub(",","",x)})
        IS <- as.data.frame(apply(IS,2,as.numeric))
        rownames(IS) <- paste(names_row)
        temp1 <- IS
        url <- "https://finance.yahoo.com/quote/"
        url <- paste0(url,stock[i],"/balance-sheet?p=",stock[i])
        wahis.session <- html_session(url)
        p <-    wahis.session %>%
          html_nodes(xpath = '//*[@id="Col1-1-Financials-Proxy"]/section/div[3]/table')%>%
          html_table(fill = TRUE)
        BS <- p[[1]]
        colnames(BS) <- BS[1,]
        BS <- BS[-c(1,2,17,28),]
        names_row <- BS[,1]
        BS <- BS[,-1] 
        BS <- apply(BS,2,function(x){gsub(",","",x)})
        BS <- as.data.frame(apply(BS,2,as.numeric))
        rownames(BS) <- paste(names_row)
        temp2 <- BS
        url <- "https://finance.yahoo.com/quote/"
        url <- paste0(url,stock[i],"/cash-flow?p=",stock[i])
        wahis.session <- html_session(url)
        p <-    wahis.session %>%
          html_nodes(xpath = '//*[@id="Col1-1-Financials-Proxy"]/section/div[3]/table')%>%
          html_table(fill = TRUE)
        CF <- p[[1]]
        colnames(CF) <- CF[1,]
        CF <- CF[-c(1,3,11,16),]
        names_row <- CF[,1]
        CF <- CF[,-1] 
        CF <- apply(CF,2,function(x){gsub(",","",x)})
        CF <- as.data.frame(apply(CF,2,as.numeric))
        rownames(CF) <- paste(names_row)
        temp3 <- CF
        assign(paste0(stock[i],'.f'),value = list(IS = temp1,BS = temp2,CF = temp3),envir = parent.frame())

      },
      error = function(cond){
        message(stock[i], "Give error ",cond)
      }
    )
  }
}

################## END OF FUNCTION ######################
#
#########################################################

symbols <- c("HOG", "GOOG", "GE")

我可以使用以下getFin(symbols)下载财务报表。但是我遇到了一个问题,因为我想使用以前的数据;

tickers <- new.env()
lapply(symbols, getFin, env=tickers)
BS <- data.frame(lapply(tickers, function(x) {viewFinancials(x, type= 'BS', period = 'A')}))
IS <- data.frame(lapply(tickers, function(x) {viewFinancials(x, type= 'IS', period = 'A')}))
CF <- data.frame(lapply(tickers, function(x) {viewFinancials(x, type= 'CF', period = 'A')}))

我在lapply(symbols, getFin, env=tickers)遇到错误,但不太确定为什么,因为我将数据以与以前相同的格式输入。

编辑:

大致了解我要实现的目标(作为最终目标)。我想要3个分别用于BS-资产负债表,IS-损益表和CF-现金流量`的数据框,每个公司列为一列。

0 个答案:

没有答案