variofit: Function that fits a model to a variogram.
In geo: Draw and Annotate Maps, Especially Charts of the North Atlantic

Description Usage Arguments Value Side Effects Details See Also

Function that fits a model to a variogram. The fitting occurs either automatically or interactively. The function is called after the function variogram which calculates the variogram to which the model is fitted. Currently only spherical model is supported. Later other models will be added.

1	variofit(vagram, model=1, option=2, interactivt=F, sill=0)

`vagram`	List with the calculated variogram. Components of the list are: \$dist mean distance of the interval. <s-example> \$vario calculated value of the variogram. \$number number of datapoints in the interval. </s-example> In nearly all cases vagram will be the output from the program variogram.
`model`	Type of model. Default is spherical. It is currently the only model supported.
`option`	Method to use in automatic fitting. Allowed values are 1,2,3 and 4. Default value is 2. For further information see below.
`interactivt`	If T the fitting is done interactively by plotting the variogram on the screen and asking the user to select sill, range and nugget by the locator function.
`sill`	Sill of the variogram, or: Limit of the variogram tending to infinity lag distances (wikipedia).

<s-example> A list with the following components. \$nugget : Estimated nugget effect \$sill : Estimated sill \$range : Estimated range \$dist : mean distance of the interval. \$vario : calculated value of the variogram. \$number : number of datapoints in the interval. </s-example>

No side effects. X11 window has to be open when the program is called.

<s-example> With automatic fitting the program fits a spherical variogram to the data by one of 4 different method. All of the methods are based on fitting the variogram by the "supsmu" of Splus that is a robust smoother. When the "supsmu" function is used it should not matter which method is used to calculate the variogram, the "supsmu" function handles outliers. The 4 method for fitting are:

1: Sill = maximum of smoothed variogram. Range the corresponding distance. 2: Sill = first maximum of smoothed variogram. Range the corresponding distance. 3: Sill = Second maximum of smoothed variogram. Range the corresponding distance. 4: Sill given. Range is the distance when the smoothed variogram reaches the sill. Good choice of sill can be var(z)

2 is the default value but of these method 2 and 4 seem the most useful. The user can plot the variogram and the fitted curve by the function plvar. Then the fitted curve can be changed by changing the components of the list returned by the program.

When using the program the user should keep in mind that only two parameters determine the interpolated values, i.e. range and nugget/sill. The absolute value of the sill is only important if the estimation variance is to be found. </s-example>

variogram, pointkriging.

geo documentation built on May 2, 2019, 5:22 p.m.