[R] Asking help about drawing and saving candle chart automatically....

박상규 birdfire94 at naver.com
Sun Oct 14 18:12:42 CEST 2012


I have trouble with saving and showing candlestick graph.
I want to draw daily 1-minute candle chart per each day and save it
The data I saved and is used in my code is as the following format:
Index,Open,High,Low,Close # header
2011-11-01  9:00:00 ,248.50,248.95,248.20,248.70
2011-11-01 15:15:00 ,250.55,250.55,250.55,250.55
2011-11-02  9:00:00 ,245.55,246.25,245.40,245.80
2011-11-04 15:15:00 ,252.50,252.50,252.50,252.50

(Some daily data does not exists sometimes. Because it was holyday.)

2011-11-07  9:00:00 ,253.00,253.10,252.60,252.80

Here is the full source that I coded for the loading csv, drawing chart,
and saving chart as jpeg file.


aa <- as.xts(read.zoo("E://log.txt", sep=",", FUN=as.chron, header=TRUE))

# convert start date of csv file as julian date
juldate_start <- as.integer(as.Date("2011-11-01"))

for(i in 1:5)
	# increased date and convert it as Date class again
	convdate <- as.Date(juldate_start + i)

	# reformat date as recognizable string by quantmod 
	format(convdate, format="yyyy-mm-dd")
	strdate <- as.character(convdate)

	# make daily range
	daterange <- paste( strdate, "::", strdate, sep="" )

	jpegname <- paste(strdate, ".jpeg", sep="")
	#### jpeg( filename=jpegname, width=1200, height=400 )
	tryCatch(chartSeries(aa, subset=daterange, type='candlesticks',

	#saveChart( 'jpeg' )
	#### dev.off()

	# wait for user input
	cat("\n","Enter x","\n") # prompt

The problems are as the follows:

1. I tried to save chart as the format like '2011-11-02.jpg', '2011-11-
03.jpg', etc. whenever count variable 'i' increases.
But I failed it. I couldn't find how to change name of char file
Could you fix my code or give me advice ?

2. The above source code showed daily chart successfully until date of 2011-
But it failed to draw chart of 2011-11-07.
Because code tried to find data in 2011-11-05, but it does not exists in
the file 'log.txt'.
How can I handle this problem ? 

Thank you in advance.

SK Park

More information about the R-help mailing list