[R-SIG-Finance] Cointegration

[Joshua Ulrich]

coef(m)[1] is the intercept in your model. coef(m)[2] is the slope.
If you wanted to run a regression with no intercept, include "+0" or
"-1" in your formula (as it says in ?lm).
--
Joshua Ulrich  |  about.me/joshuaulrich
FOSS Trading  |  www.fosstrading.com


On Wed, Aug 7, 2013 at 1:19 PM, Josef Jauk <josef.jauk at live.at> wrote:
> I have build a cointegration model to select stocks for pairs trading. My problem with it is to refer within a loop to a list of symbols. While the code runs without generating any warnings I get the wrong results:
>
> e.g. for the combination BP ~ AMD I get beta=36.45182 and p-value=0.241386 while it should be beta=5.153268 and p-value=0.1557386
>
> Thanks for every help to solve the problem.
>
> the code is:
>
> #Cointegration Combinations
> require(quantmod)
> require(zoo)
> require(quadprog)
> require(tseries)
>
> setInternet2()
>
> symList <- c('MSN','GOOG','YHOO', 'BA', 'SI', 'BP', 'AMD')
>
> xtsStock <- xts()
> xtsFinalPrice <- xts()
>
> for (i in 1:length(symList)) {
>    xtsStock <- getSymbols(symList[i], auto.assign=FALSE)
>    xtsFinalPrice <- merge(xtsFinalPrice, xtsStock[,6]) #get the adjusted close price
> }
>
> colnames(xtsFinalPrice) <- symList
> t <- as.data.frame(xtsFinalPrice)
> comb <- combn(symList, 2, FUN=NULL, simplify=TRUE)
> ergebnis <- cbind(t(comb), 0, 0)
> colnames(ergebnis) <- c("St.Y", "St.X", "BETA", "p-value")
>
> for (i in 1:((length(symList) * (length(symList)-1)) / 2)) {
>    #m <- lm(MSN ~ AMD + 0, data=t)
>    m <- lm(paste(comb[1,i],"~",comb[2,i],sep=" ") , data=t)
>    beta <- coef(m)[1]
>    #sprd <- t$MSN - beta * t$AMD
>    sprd <- get(comb[1,i], pos=t) - beta * get(comb[2,i], pos=t)
>    ht <- adf.test(sprd, alternative="stationary", k=trunc((length(sprd)-1)^(1/3)))
>    ergebnis[i,3] <- as.numeric(beta)
>    ergebnis[i,4] <- as.numeric(ht$p.value)
> }
>
> Josef
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> R-SIG-Finance at 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.