| Type: | Package |
| Title: | ICSS Algorithm by Inclan/Tiao (1994) |
| Version: | 1.1 |
| Date: | 2021-04-22 |
| Maintainer: | Siegfried Köstlmeier <siegfried.koestlmeier@gmail.com> |
| Description: | The Iterative Cumulative Sum of Squares (ICSS) algorithm by Inclan/Tiao (1994) https://www.jstor.org/stable/2290916 detects multiple change points, i.e. structural break points, in the variance of a sequence of independent observations. For series of moderate size (i.e. 200 observations and beyond), the ICSS algorithm offers results comparable to those obtained by a Bayesian approach or by likelihood ration tests, without the heavy computational burden required by these approaches. |
| License: | GPL-2 |
| Depends: | R (≥ 3.5.0) |
| Imports: | rstack |
| Suggests: | testthat |
| Encoding: | UTF-8 |
| LazyData: | true |
| NeedsCompilation: | no |
| Packaged: | 2021-04-22 13:06:47 UTC; LocalAdmin |
| Author: | Siegfried Köstlmeier
 [aut, cre,
cph] |
| Repository: | CRAN |
| Date/Publication: | 2021-04-22 14:00:19 UTC |
Iterative Cumulative Sum of Squares (ICSS)
Description
ICSS implements the Iterative Cumulative Sum of Squares (ICSS) algorithm by Inclan/Tiao (1994).
The test detects structural breakpoints in the variance of time series data.
Usage
ICSS(data, demean = FALSE)
Arguments
data |
A numerical vector |
demean |
An object of class |
Value
ICSS returns a numerical vector containing the location of structural breakpoints or NA if none breakpoints are found.
References
Inclan, C., & Tiao, G. C. (1994): Use of cumulative sums of squares for retrospective detection of changes of variance. Journal of the American Statistical Association, 89(427), 913-923. https://www.jstor.org/stable/2290916.
Examples
## load demo data
data(data)
breakpoints <- ICSS(data)
Sample data for Inclan/Tiao (1994)
Description
Generated random data (n=700) with following the scheme in Inclan/Tiao (1994):
[1;390]Mean: 0; Variance: 1.000
[391;517]Mean: 0; Variance: 0.365
[518;700]Mean: 0; Variance: 1.033
Usage
data(data)
Examples
## load data
data(data)
## calculate the variance until the first breakpoint.
data_var <- var(data[1:390])