CODEWITHSUNDEEP
rlinear-regression
Cyber Student
Cyber Student

Reputation: 311

Linear Regression Curve in R

I am trying to implement the linear Regression curve mentioned at this link in R and need help.

Link: Linear Regression Curve

I found the following ThinkScript code that implements what I am looking for. Any help in converting it to R?

script inertiaTS {
    input y = close;
    input n = 20;
    rec x = x[1] + 1;
    def a = (n * Sum(x * y, n) - Sum(x, n) * Sum(y, n) ) / ( n * Sum(Sqr(x), n) - Sqr(Sum(x, n)));
    def b = (Sum(Sqr(x), n) * Sum(y, n) - Sum(x, n) * Sum(x * y, n) ) / ( n * Sum(Sqr(x), n) - Sqr(Sum(x, n)));
    plot InertiaTS = a * x + b;
}

Here is what I have so far..

Sqr <- function(x) {
return (x^2)
}

inertiaTS <- function(y, n) {
    x <- x + 1;
    a <- (n * rollapply( x*y, n, sum) - rollapply( x, n, sum ) * rollapply( y, n, sum )) / ( n * rollapply( Sqr(x), n, sum ) - Sqr(rollapply( x, n, sum )))
    b <- (rollapply( Sqr(x), n, sum) * rollapply( y, n, sum ) - rollapply( x, n, sum ) * rollapply( x*y, n, sum ) ) / ( n * rollapply( Sqr(x), n, sum ) - Sqr

(rollapply( x, n, sum )))
    return (a * x + b)
}

When I make a call to the function with

lrc <- inertiaTS(Cl(stockData$AAPL), 20)

I get the following error. Any help with this?

Error in seq.default(start.at, NROW(data), by = by) : wrong sign in 'by' argument

Thanks.

Upvotes: 1

Views: 1233

Answers (1)

Robert
Robert

Reputation: 5152

The rollapply and lm do the task:

library(zoo)
inertiaTS <- function(y, n) {
  x <- 1:n;
  c.ab=rollapply(y,n,function(yt){
    coef(lm(yt~x))
  },align = "right")                                                                                                                    
  plot(y,col=2)
  lines(c.ab[ ,2]*x[n]+c.ab[ ,1],col=4,lwd=2)
  list(axpb=c.ab[ ,2]*x[n]+c.ab[ ,1],rolcoef=c.ab)
}

#simulate any ts series
set.seed(123)
y=ts(10+sort(rnorm(150,sd=.8))+runif(150))
plot(y)
res<-inertiaTS(y,20)

enter image description here tail(res$axpb)

[1] 11.84216 11.95979 12.06342 12.05539 12.06578 12.19312

Upvotes: 1

Related Questions