fractalBlock: Class constructor for block-dependent estimators for...
In fractal: A Fractal Time Series Modeling and Analysis Package

Description Usage Arguments S3 METHODS See Also Examples

Class constructor for fractalBlock.

1
2
3

fractalBlock(domain,estimator, exponent, exponent.name,
    scale, stat, stat.name, detrend, overlap,
    data.name, sum.order, series, logfit, sdf=NULL)

`domain`	character string defining the domain in which the calculations took place, e.g., in the time or frequency domain.
`estimator`	character string briefly describing the estimator.
`exponent`	numerical value representing the scaling exponent.
`exponent.name`	character string defining the name of the scaling exponent.
`scale`	numeric vector containing the scales used in the analysis.
`stat`	numeric vector containing the statistic calculated in the analysis.
`stat.name`	character vector describing the name for the calculated statistic.
`detrend`	character string describing any series detrending used as a preprocessive measure. `NULL` values are allowed and signify no detrending.
`overlap`	numeric value on [0,1] defining the fraction of overlap used in adjacent blocks of data during the aggregation process.
`data.name`	character string defining the name of the input series.
`sum.order`	integer representing the sum order.
`series`	a numeric vector containing the input series.
`logfit`	a linear regression model (such as that output by `lm`, `lmsreg`, or `ltsreg`) containing the regression model of the `log(scale)` versus `log(stat)` data.
`sdf`	spectral density function. Default: `NULL`.

eda.plot

extended data analyis plot of the data. Available options are:

cex: character expansion ala par. Default: 1.
col: line color index ala par. Default: 2.

plot

plots a summary of the results. Available options are:

pch: plot character ala par. Default: 18.
col: color index ala par for a plot of the data. the first two elements are used to color the data and the regression line, respectively. Default: c(1,8).
lty: line types (ala par) for the data and regression line plot, respectively. Default: c(1,1).
grid: list of grid objects whose default values are grid=list(lty=2, col=16, density=3), key=TRUE, add=FALSE, cex=1, ...).
...: Additional plot arguments (set internally by the par function).

print

prints the object. Available options are:

justify: text justification ala prettPrintList. Default: "left".
sep: header separator ala prettyPrintList. Default: ":".
n.digits: number of digits ala prettyPrintList. Default: 5.
...: Additional print arguments sent directly to the prettyPrintList function).

hurstBlock, DFA.

## construct a fractalBlock object 
xvar <- 2^(1:10)
yvar <- 0.3^(1:10)
z <- fractalBlock(domain="time", estimator="estimator", exponent=pi, exponent.name="PI",
    scale=xvar, stat=yvar, stat.name="My Stat",
    detrend=NULL, overlap=0.2, data.name="My Series",
    sum.order=-1, series=rnorm(2^10),
    logfit=lm(y ~ x, data=data.frame(x=log(xvar), y=log(yvar))))

## print the result 
print(z)

## plot the result 
plot(z)

Loading required package: splus2R
Loading required package: ifultools
Estimator for My Series
-----------------------
PI estimate            : 3.141593 
Domain                 : Time 
Statistic              : My Stat 
Length of series       : 1024 
Block overlap fraction : 0.2 
Scale ratio            : 2 
Preprocessing          : 1st order difference 
                                                                               
Scale   2.0 4.00 8.000 16.0000 32.00000 6.40e+01 1.280e+02 2.560e+02 5.1200e+02
My Stat 0.3 0.09 0.027  0.0081  0.00243 7.29e-04 2.187e-04 6.561e-05 1.9683e-05
                  
Scale   1.0240e+03
My Stat 5.9049e-06