s-news
[Top] [All Lists]

Re: TimeSeries Data Manipulation

To: "Chen,Sichong" <sichong.chen@postgrad.manchester.ac.uk>
Subject: Re: TimeSeries Data Manipulation
From: Tony Plate <tplate@blackmesacapital.com>
Date: Tue, 25 Apr 2006 18:32:47 -0600
Cc: s-news@lists.biostat.wustl.edu
In-reply-to: <20060425234547.35887100955A@mailgate.biostat.wustl.edu>
References: <20060425234547.35887100955A@mailgate.biostat.wustl.edu>
User-agent: Mozilla Thunderbird 1.0.5 (Windows/20050711)
To answer your first question, use seriesData() to extract the vector or matrix of data from a timeSeries or signalSeries object.

To combine colums of multiple timeSeries, use the function seriesMerge():

> A <- timeSeries(1:3, pos=1:3)
> B <- timeSeries(cbind(B1=4:6, B2=7:9), pos=1:3)
> A
  Positions 1
 1960/01/02 1
 1960/01/03 2
 1960/01/04 3
> B
  Positions B1 B2
 1960/01/02 4  7
 1960/01/03 5  8
 1960/01/04 6  9
> seriesData(A)
[1] 1 2 3
> seriesMerge(A, B)
  Positions 1 B1 B2
 1960/01/02 1 4  7
 1960/01/03 2 5  8
 1960/01/04 3 6  9
> X <- seriesMerge(A, B)
> colIds(seriesData(X))[1] <- "A"
> X
  Positions A B1 B2
 1960/01/02 1 4  7
 1960/01/03 2 5  8
 1960/01/04 3 6  9
>


Chen,Sichong wrote:
Dear Users

I have a silly question. How can I change a Time Series vector or matrix into a non-Time Series vector or matrix?

My problem is that I have one time series vector*** A* (280*1) and one time series matrix*** B* (280*3), which has the same time period. What I want to do is to combine these two together and create a new matrix (280*4) like/// rbind///// for the no-time series dataframe. Since/// rbind///// does not work for time series data, I tried to use*** B[,4] = A****,* but fails if it is, which says “///Problem in x[, j] <- value: Array subscript (2) out of bounds, should be at most 1////”/. So I want to convert A, B back to non-time series object, combine and convert again.

Do you have some idea what’s going wrong in my problem. I though it should not be that difficult and complicated to add one more column to a time series object.

Many thanks,

Sichong

SPlus 7.0 & FinMetrics 2.0 User

Windows XP



<Prev in Thread] Current Thread [Next in Thread>