问题
I'm new to learning R and I'm having an issue with some of my R code. I placed all the code for your convenience so that you can see the logic in what I am trying to do. My issue is renaming the header of my xts object Monthly_Quotes. I understand that when having an invalid stock symbol, the getsymbols function will not retrieve the quotes for "zzzz", which is why I am running into the issue of renaming the header. I'd like to resolve this such that if I have a much larger list of ticker symbols that do not get downloaded, that I won't have an issue renaming the header from my list of ticker_symbols.
#Environment to store stock symbols data
ETF_Data <- new.env()
#Dates needed for getSymbols function
sDate <- as.Date("2007-09-04") #Start Date
eDate <- as.Date("2014-09-02") #End Date
#Assign vector of ticker symbols
ticker_symbol <- c("zzzz","IVW","JKE","QQQ","SPYG","VUG" )
#Load functions
source("Functions.R")
#Create empty list
Temp_ETF_Data <- list()
#Variable is a connection to a specific file that appends the output to the file
connection_string <- file("Log_File.txt", open = "wt")
#Sink allows the R output to a connection file which is connection_string
#Log_File.txt is created to the directory that has been set
sink(connection_string, type="message")
#Loop to retrieve prices from Yahoo's API and assign only the Adjusted column to the list.
for(i in ticker_symbol){
tryCatch(
{
getSymbols(i, env = ETF_Data, from = sDate, to = eDate, src = "yahoo", warning = FALSE)
Temp_ETF_Data[[i]] <- Ad(ETF_Data[[i]])
},
error = function(e) {
message("-ERROR-")
message(paste("Error for ticker symbol :", i, "\n"))
message("Here's the original error message: ")
message(e)
message("")
return(i)},
finally = {
message(paste("Data was processed for symbol:", "[", i, "]" ))
message("\n","******************************************", "\n")
})
}
#Turns off message once it is on the console and reads the line.
#Then it closes the sink connection and closes the connection to the Log_File.txt
sink(type = "message")
readLines("Log_File.txt")
close(connection_string)
#Merge list into columns from list
Daily_Quotes <- do.call(merge, Temp_ETF_Data)
#Create new xts object with the 1st trading day price of each month and assign column headers
Monthly_Quotes <- Daily_Quotes[startpoints(Daily_Quotes,'months')]
#This piece of code creates this error
# "Error in `colnames<-`(`*tmp*`, value = c("zzzz", "IVW", "JKE", "QQQ", :
# length of 'dimnames' [2] not equal to array extent"
names(Monthly_Quotes) <- c(ticker_symbol)
I tried this:
ticker_symbol <- colnames(Monthly_Quotes)
names(Monthly_Quotes) <- c(ticker_symbol)
回答1:
This does not directly answer your question (see the comment), but wouldn't this be easier:
library(quantmod)
ticker_symbol <- c("zzzz","IVW","JKE","QQQ","SPYG","VUG" )
sDate <- as.Date("2007-09-04") #Start Date
eDate <- as.Date("2014-09-02") #End Date
get.symbol <- function(ticker) {
tryCatch(temp <- Ad(getSymbols(ticker, auto.assign=FALSE,
from = sDate, to = eDate, warning = FALSE)),
error = function(e) {
message("-ERROR-")
message(paste("Error for ticker symbol :", ticker, "\n"))
message("Here's the original error message: ")
message(e)
message("")
return(NULL)},
finally = {
message(paste("Data was processed for symbol:", "[", ticker, "]" ))
message("\n","******************************************", "\n")
})
}
result <- do.call(cbind,lapply(ticker_symbol,get.symbol))
names(result) <- gsub("\\..+","",names(result)) # remove ".Adjusted" from names
head(result)
# IVW JKE QQQ SPYG VUG
# 2007-09-04 61.71 65.37 46.56 52.42 57.36
# 2007-09-05 61.37 64.78 46.10 51.95 56.85
# 2007-09-06 61.47 65.13 46.06 52.18 57.13
# 2007-09-07 60.58 64.06 45.21 51.29 56.31
# 2007-09-10 60.36 64.13 45.18 51.24 56.12
# 2007-09-11 61.19 65.03 45.86 51.90 56.89
This takes advantage of the auto.assign=...
argument of getSymbols(...)
to avoid all this environment-setting and accumulation the temporary results in a list.
来源:https://stackoverflow.com/questions/27489387/r-code-to-rename-header-of-an-xts-object-using-nameobject-vector