# RFfractaldim: RFfractaldimension In RandomFields: Simulation and Analysis of Random Fields

## Description

The function estimates the fractal dimension of a process

## Usage

 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14``` ```RFfractaldim(x, y = NULL, z = NULL, data, grid, bin=NULL, vario.n=5, sort=TRUE, fft.m = c(65, 86), ## in % of range of l.lambda fft.max.length=Inf, fft.max.regr=150000, fft.shift = 50, # in %; 50:WOSA; 100: no overlapping method=c("variogram", "fft"), mode = if (interactive ()) c("plot", "interactive") else "nographics", pch=16, cex=0.2, cex.main=0.85, printlevel = RFoptions()\$basic\$printlevel, height=3.5, ...) ```

## Arguments

 `x` \argX If `x` is not given and `data` is not an sp object, a grid with unit grid length is assumed `y,z` \argYz `data` the values measured; it can also be an sp object `grid` \argGrid `bin` sequence of bin boundaries for the empirical variogram `vario.n` first `vario.n` values of the empirical variogram are used for the regression fit that are not `NA`. `sort` If `TRUE` then the coordinates are permuted such that the largest grid length is in `x`-direction; this is of interest for algorithms that slice higher dimensional fields into one-dimensional sections.
 `fft.m` numeric vector of two components; interval of frequencies for which the regression should be calculated; the interval is given in percent of the range of the frequencies in log scale. `fft.max.length` The first dimension of the data is cut into pieces of length `fft.max.length`. For each piece the FFT is calculated and then the average for all pieces is taken. The pieces may overlap, see the argument `fft.shift`. `fft.max.regr` If the `fft.m` is too large, parts of the regression fit will take a very long time. Therefore, the regression fit is calculated only if the number points given by `fft.m` is less than `fft.max.regr`. `fft.shift` This argument is given in percent [of `fft.max.length`] and defines the overlap of the pieces defined by `fft.max.length`. If `fft.shift=50` the WOSA estimator is given; if `fft.shift=100` no overlap exists. `method` list of implemented methods to calculate the fractal dimension; see Details `mode` character. A vector with components `'nographics'`, `'plot'` or `'interactive'`: `'nographics'`no graphical output `'plot'`the regression line is plotted `'interactive'`the regression domain can be chosen interactively Usually only one mode is given. Two modes may make sense in the combination `c("plot", "interactive")`. In this case, all the results are plotted first, and then the interactive mode is called. In the interactive mode, the regression domain is chosen by two mouse clicks with the left mouse; a right mouse click leaves the plot. `pch` vector or scalar; sign by which data are plotted. `cex` vector or scalar; size of `pch`. `cex.main` The size of the title in the regression plots. `printlevel` integer. If `printlevel` is 0 nothing is printed. If `printlevel=1` error messages are printed. If `printlevel=2` warnings and the regression results are given. If `printlevel>2` tracing information is given. `height` height of the graphics window `...` graphical arguments

## Details

The function calculates the fractal dimension by various methods:

• variogram method

• Fourier transform

## Value

The function returns a list with elements `vario`, `fft` corresponding to the 2 methods given in the Details.

Each of the elements is itself a list that contains the following elements.

 `x` the x-coordinates used for the regression fit `y` the y-coordinates used for the regression fit `regr` the return list of the `lm`. `sm` smoothed curve through the (x,y) points `x.u` `NULL` or the restricted x-coordinates given by the user in the interactive plot `y.u` `NULL` or y-coordinates according to `x.u` `regr.u` `NULL` or the return list of `lm` for `x.u` and `y.u` `D` the fractal dimension `D.u` `NULL` or the fractal dimension corresponding to the user's regression line

## References

variogram method

• Constantine, A.G. and Hall, P. (1994) Characterizing surface smoothness via estimation of effective fractal dimension. J. R. Statist. Soc. Ser. B 56, 97-113.

fft

• Chan, Hall and Poskitt (1995)

`RMmodel`, `RFhurst`
 ```1 2 3 4 5 6``` ```RFoptions(seed=0) ## *ANY* simulation will have the random seed 0; set ## RFoptions(seed=NA) to make them all random again x <- seq(0, 10, 0.001) z <- RFsimulate(RMexp(), x) RFfractaldim(data=z) ```