rbf.tcv: table of rbf cross validation, leave-one-out

Description Usage Arguments Details Value See Also Examples

Description

Generates a table with the results of the evaluation of radial basis functions (rbf): gaussian (GAU), exponential (EXPON), trigonometric (TRI), thin plate spline (TPS), completely regularized spline (CRS), spline with tension (ST), inverse multiquadratic (IM), and multiquadratic (M) from the leave-one-out cross validation method.

Usage

1
rbf.tcv(formula, data, eta, rho, n.neigh, func)

Arguments

formula

formula that defines the dependent variable as a linear model of independent variables; suppose the dependent variable has name z, for a rbf detrended use z~1, for a rbf with trend, suppose z is linearly dependent on x and y, use the formula z~x+y (linear trend).

data

SpatialPointsDataFrame: should contain the dependent variable, independent variables, and coordinates.

eta

the optimal smoothing parameter; we recommend using the parameter found by minimizing the root-mean-square prediction errors using cross-validation

rho

value of optimal parameter robustness; we recommend using the parameter found by minimizing the root-mean-square prediction errors using cross-validation. eta and rho parameters can be optimized simultaneously, through the bobyqa function from nloptr or minqa packages

n.neigh

number of nearest observations that should be used for a rbf prediction, where nearest is defined in terms of the spatial locations

func

radial basis function model type, e.g. "GAU", "EXPON", "TRI", "TPS", "CRS", "ST", "MI" and "M", are currently available

Details

Leave-one-out cross validation (LOOCV) visits a data point, predicts the value at that location by leaving out the observed value, and proceeds with the next data point. The observed value is left out because rbf would otherwise predict the value itself.

Value

data frame contain the data coordinates, prediction columns, observed values, residuals, the prediction variance, zscore (residual divided by standard error) which left with NA's, and the fold column which is associated to cross-validation count. Prediction columns and residuals are obtained from cross-validation data points

See Also

rbf

Examples

1
2
3
data(preci)
coordinates(preci)<-~x+y
rbf.tcv(prec~x+y, preci, eta=0.1460814, rho=0, n.neigh=9, func="TPS")

Example output

Loading required package: gstat
Loading required package: genalg
Loading required package: MASS
Loading required package: sp
Loading required package: minqa

  |                                                                            
  |                                                                      |   0%
  |                                                                            
  |=======                                                               |  10%
  |                                                                            
  |==============                                                        |  20%
  |                                                                            
  |=====================                                                 |  30%
  |                                                                            
  |============================                                          |  40%
  |                                                                            
  |===================================                                   |  50%
  |                                                                            
  |==========================================                            |  60%
  |                                                                            
  |=================================================                     |  70%
  |                                                                            
  |========================================================              |  80%
  |                                                                            
  |===============================================================       |  90%
  |                                                                            
  |======================================================================| 100%
   var1.pred var1.var observed    residual zscore fold x y
1   398.3808       NA      420  21.6191767     NA    1 1 4
2   412.5996       NA      410  -2.5996082     NA    2 1 2
3   420.0921       NA      405 -15.0921052     NA    3 3 3
4   419.1315       NA      415  -4.1315291     NA    4 3 0
5   429.4019       NA      430   0.5981334     NA    5 5 1
6   418.6864       NA      425   6.3136167     NA    6 5 3
7   422.8531       NA      415  -7.8530624     NA    7 6 4
8   433.6800       NA      435   1.3200272     NA    8 6 1
9   424.7823       NA      425   0.2177269     NA    9 6 3
10  431.4525       NA      430  -1.4524893     NA   10 7 2

geospt documentation built on May 2, 2019, 4:51 p.m.