pfCircular: Circular block bootstrap procedure applied to charcoal...

Description Usage Arguments Value Author(s) References Examples

View source: R/pfCircular.R

Description

Block bootstrap has been proposed to test the significances of changes in stationary time series (Kunsch 1989). This procedure consists of splitting each charcoal series into n-b+1 overlapping blocks of data, where n is sample size and b the block size. These blocks are used to reconstruct resampled individual charcoal series that are in turn used to estimate the confidence intervals around the charcoal series composite mean.

Usage

1
2
pfCircular(comp, b = NULL, conf = c(0.05, 0.95), nboot = 1000,
  AgeLim = NULL)

Arguments

comp

A "pfComposite" object

b

A numeric giving block size, if NULL the optimal block size for a given series is given by: b= 2x(-1 /log(p)), where p is the lag one autocorrelation coefficient of that series (Adams, Mann & Ammann 2003).

conf

Numeric, calculated confidence intervals.

nboot

Numeric, number of bootstrap replicates.

AgeLim

Numeric, years defining a period to restrict the analysis to.

Value

out

A "pfCircular" object with estimated confidence intervals.

Author(s)

O. Blarquez

References

Kunsch, H. R. 1989. The jackknife and the bootstrap for general stationary observation s. The Annals of Statistics 17:1217-1241.

Adams, J. B., M. E. Mann, and C. M. Ammann. 2003. Proxy evidence for an El Nino-like response to volcanic forcing. Nature 426:274-278.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
## Not run: 
ID=pfSiteSel(lat>49, lat<75, long>6, long<50)
plot(ID,zoom="world")
TR1=pfTransform(ID, method=c("MinMax","Box-Cox","Z-Score"),BasePeriod=c(200,2000))

## Circular block bootstrapp

COMP=pfComposite(TR1, binning=TRUE, bins=seq(0,2000,100))
circ=pfCircular(COMP,conf=c(0.005,0.025,0.975,0.995),nboot=100)
plot(circ)

## End(Not run)

oblarquez/paleofire documentation built on Oct. 19, 2018, 1:09 p.m.