stockreturns22=data.frame(ret22=(data22)[,4],
                          timestamps=(data22)[,2],format="%Y/%m/%d")
stockreturns22.ts= xts(stockreturns22$ret22,
order.by=as.Date(stockreturns22$timestamps))
rt22=ts(stockreturns22$ret22)
tt22=length(rt22)
lm22=lm(rt22[-1]~rt22[-tt22]+rt22[-tt22]:I(rt22[-tt22]<0))
summary(lm22)

tauseq=c(0.02,0.1,0.2,0.5,0.8,0.9,0.98)
qslope22=NULL; qr22=list()
qqslope22=NULL;
for(i in 1:length(tauseq)){

qr22[[i]]=rq(rt22[-1]~rt22[-tt22]+rt22[-tt22]:I(rt22[-tt22]
<0),tau=tauseq[i])
forecast(qr22)
qslope22[i]=qr22[[i]]$coef[2]
qqslope22[i]=qr22[[i]]$coef[3]
}

Coefficients: (1 not defined because of singularities)
                                     Estimate Std. Error t value Pr(>|t|)

(Intercept)                         504.65017   14.84839   33.99   <2e-16
***
rt22[-tt22]                           0.34867    0.01603   21.75   <2e-16
***
rt22[-tt22]:I(rt22[-tt22] <= 0)TRUE        NA         NA      NA       NA

---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Error in rq.fit.br(x, y, tau = tau, ...) : Singular design matrix