[prev in list] [next in list] [prev in thread] [next in thread] 

List:       r-sig-robust
Subject:    Re: [RsR] Time-series outlier problem
From:       Victor Yohai <victoryohai () gmail ! com>
Date:       2021-06-14 19:58:17
Message-ID: 007301d76157$9f2c2b40$dd8481c0$ () uba ! ar
[Download RAW message or body]

Dear Ajay

A robust way to  find outliers in a time series is using the function
arima.rob of the R package robustarima.  Details on how  this  function is
used can be found in its help. If you want to know more about the procedure
used for this function you can consult Chapter 8 of Robust Statistics:
Theory and methods (with R) of Maronna, Martin, Salibian-Barrera and Yohai.

Víctor Yohai     



-----Mensaje original-----
De: R-SIG-Robust [mailto:r-sig-robust-bounces@r-project.org] En nombre de
Ajay Shah
Enviado el: viernes, 11 de junio de 2021 6:56
Para: r-sig-robust@r-project.org
Asunto: [RsR] Time-series outlier problem

We've been using forecast::tsoutliers. In many situations it's nice, but I
have a MWE where it seems to get confused:

library(forecast)
x <- sin(seq(0,6.28,0.01)) # There's no noise

tsoutliers(ts(x)) # nothing is amiss.

x[4]  <- 2
tsoutliers(ts(x)) ## Weird

x[4]  <-  100
tsoutliers(ts(x)) ## Weird

It shows a lot of outliers and offers bizarre replacements.

We are working with time-series of about N=100 with gentle time series
structure. Do folks here have suggestions on how best to handle outlier
detection and correction for this?

-- 
Ajay Shah
ajayshah@mayin.org
http://www.mayin.org/ajayshah

	[[alternative HTML version deleted]]

_______________________________________________
R-SIG-Robust@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-robust

_______________________________________________
R-SIG-Robust@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-robust
[prev in list] [next in list] [prev in thread] [next in thread]