Hi
I am getting this error even though I'm pretty sure I am not out of bounds:
Error in `[.xts`(one, trim:length(two), ) : subscript out of bounds
In addition: Warning message:
In trim:length(two) :
numerical expression has 305 elements: only the first used
I have tried all sorts of things but clearly I am not figuring out what
is going wrong.
I am developing code to recover prices from IB and test for cointegration.
There are two files attached:
* checkPairFromIB
* metals.csv
It all worked fine until I discovered gold and silver returned data of
different lengths: Gold had 320 days of data and silver 305. The calls
for data at lines 25 and 26 are identical so I don't know why. Can
anyone shed light on this?
I then tried to trim the longer table by disposing of the earlier 15
lines (see lines 44 to 55 of the code). This is where the error
messages came up. Can anyone point out what I am doing wrong?
Thanks.
Stephen
--
-----------------------------------------------------------------------------------------------------------------------------------
Stephen Choularton PhD, FIoD
0413 545 182
# this is code to check the commodities for cointegration and produce a hedge
ratio
# need to create a loop to apply the function to each pair in the file
# IBrokersRef()
library("IBrokers")
library("urca") # for johansen
library(zoo)
library(tseries)
tws <- twsConnect()
reqCurrentTime(tws)
checkPairFromIB <- function(sym1, mkt1, sym2, mkt2) {
contract1 <- twsContract(0,sym1, "CONTFUT",
mkt1,"","","0.0","USD","","","",NULL,NULL,"0")
contract2 <- twsContract(0,sym2, "CONTFUT",
mkt2,"","","0.0","USD","","","",NULL,NULL,"0")
#http://www.sthda.com/english/wiki/subsetting-data-frame-rows-in-r
#print(contract1)
one <- reqHistoricalData(tws, contract1, endDateTime =
format(Sys.Date(), "%Y%m%d 24:00:00"), duration = "3 M")
two <- reqHistoricalData(tws, contract2, endDateTime =
format(Sys.Date(), "%Y%m%d 24:00:00"), duration = "3 M")
one <- one[,1:5]
two <- two[,1:5]
# Give columns more usable names
colnames(one) <- c('Open', 'High', 'Low', 'Close', 'Volume')
colnames(two) <- c('Open', 'High', 'Low', 'Close', 'Volume')
#print(one)
print(length(one))
#print(one)
print(length(two))
#print(two)
# trim the longer one
if (length(one) > length(two)){
trim <- one - two
one <- one[trim:length(two),]
} else if (length(one) < length(two)){
trim <- two - one
two <- two[trim:length(one),]
}
#print(one)
print(length(one))
#print(one)
print(length(two))
#print(two)
# failsafe if some market has closed
if (length(one) == length(two)){
jotest=ca.jo(data.frame(one$Close,two$Close), type="trace",
K=2, ecdet="none", spec="longrun")
#print(summary(jotest))
#print("10pc")
#print(jotest@cval[2,1])
#print("test")
#print(jotest@teststat[2])
#print("hedge ratio")
#print(jotest@V[2,1])
if (jotest@V[2] > jotest@cval[2,1]){
print(c(sym1, sym2, "cointegrated", "hedge",
jotest@V[2,1] ))
} else {
print(c(sym1, sym2, "not cointegrated"))
}
} else {
print("data of different lengths")
}
#https://stackoverflow.com/questions/42741633/extract-information-from-rs-johansen-procedure-ca-jo-summary
#http://r.789695.n4.nabble.com/How-to-automatically-extract-test-result-from-Johansen-test-td4519468.html
#https://cran.r-project.org/web/packages/urca/urca.pdf
}
#lets get the symbols
symbols <- read.csv("C:\\Users\\Stephen\\Documents\\R\\2018 June\\metals.csv",
stringsAsFactors=F)
print(symbols)
print ('##################')
#print(nrow(symbols))
seq(1, nrow(symbols), by = 2)
for (i in seq(1, nrow(symbols) - 1, by = 1)) {
for (j in seq(i + 1, nrow(symbols), by = 1)) {
print(c(i, j))
checkPairFromIB(symbols[i,1],
symbols[i,2],symbols[j,1],symbols[j,2])
print("###################")
#print(symbols[i,1])
#print(symbols[i,2])
#print(symbols[j,1])
#print(symbols[j,2])
print("pacing limit 10 seconds")
Sys.sleep(10)
}
}
#gold, silver
#checkPairFromIB("GC", "NYMEX","SI","NYMEX")
twsDisconnect(tws)
#head(one)
#head(two)
#pacing limit
#Sys.sleep(10)
sym,mkt
GC,NYMEX
SI,NYMEX
PL,NYMEX
PA,NYMEX
_______________________________________________
R-SIG-Finance@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only. If you want to post, subscribe first.
-- Also note that this is not the r-help list where general R questions should
go.
