R help - Dec 2011 - combining arima and ar function

Hi everyone

 

I've got a problem regarding the arima() and the ar() function for
autoregressive series. I would simply like to combine them. 

 

To better understand my question, I first show you how I'm using these two
functions individually (see file in the attachement).

 

1)      apply(TSX,2, function(x) ar(na.omit(x),method="mle")$order
# this function finds the optimal (autoregressive) order for each series
(briefly speaking: for each series I get a specific number)

2)      apply(TSX,2,function(x)arima(x,order=c(1,0,0))$residuals
# this function puts an autoregressive model of order 1 on each series

 

 

What I'd like to do, is to combine them, such that the resulting orders
(numbers) from the first function are used as orders in the second function.
So in the specific example attached the results from the first function are
6,5,1 and these numbers should be used as the order in function two. I tried
to simply put the two functions together as follows:

 

apply(TSX,2,function(x) arima(x,order=c((apply(TSX,2,
function(x)ar(na.omit(x),method="mle")$order))),0,0))

 

However, I get the error message " 'order' must be a non-negative
numeric
vector of length 3".

 

Any hints? I hope you can help me with this issue.

 

Many thanks and best regards, S.B.

I think your attachment got scrubbed so I can't verify this, but i
think the difficulty is that the "inner" apply returns a whole set of
orders, instead of just one and that throws arima off: would this
work?

getOrder <- function(x) ar(na.omit(x), method = "mle")$order

all.equal(apply(TSX, 2, getOrder), apply(TSX,2, function(x)
ar(na.omit(x),method="mle")$order) # Verify it works.

apply(TSX,2,function(x)arima(x,order=c(getOrder(x),0,0))$residuals

Michael

On Thu, Dec 1, 2011 at 4:57 AM, Samir Benzerfa <benzerfa at gmx.ch>
wrote:
> Hi everyone
>
>
>
> I've got a problem regarding the arima() and the ar() function for
> autoregressive series. I would simply like to combine them.
>
>
>
> To better understand my question, I first show you how I'm using these
two
> functions individually (see file in the attachement).
>
>
>
> 1) ? ? ?apply(TSX,2, function(x)
ar(na.omit(x),method="mle")$order
> # this function finds the optimal (autoregressive) order for each series
> (briefly speaking: for each series I get a specific number)
>
> 2) ? ? ?apply(TSX,2,function(x)arima(x,order=c(1,0,0))$residuals
> # this function puts an autoregressive model of order 1 on each series
>
>
>
>
>
> What I'd like to do, is to combine them, such that the resulting orders
> (numbers) from the first function are used as orders in the second
function.
> So in the specific example attached the results from the first function are
> 6,5,1 and these numbers should be used as the order in function two. I
tried
> to simply put the two functions together as follows:
>
>
>
> apply(TSX,2,function(x) arima(x,order=c((apply(TSX,2,
> function(x)ar(na.omit(x),method="mle")$order))),0,0))
>
>
>
> However, I get the error message " 'order' must be a
non-negative numeric
> vector of length 3".
>
>
>
> Any hints? I hope you can help me with this issue.
>
>
>
> Many thanks and best regards, S.B.
>
>
>
>
>
>
>
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>