predict.slmfit: Perform Finite Population Block Kriging

In highamm/FPBK-with-Detection: Finite Population Block Kriging with Detection

Description

Takes the output of slmfit and uses FPBK to predict the counts on the unsampled sites. The column with the counts should have numeric values for the observed counts on the sampled sites and 'NA' for any site that was not sampled.

Usage

## S3 method for class 'slmfit'
predict(object, FPBKcol = NULL, detinfo = c(1, 0),
  ...)

Arguments

object	is an object generated from slmfit
FPBKcol	is the name of the column that contains the weights for prediction. The default setting predicts the population total.
detinfo	is a vector consisting of the mean detection probability and its standard error. By default, it is set to c(1, 0), indicating perfect detection (1) with no variance (0). Using the default assumes perfect detection.
...	Additional arguments

Value

a list of class sptotalPredOut with

• the estimated population total

• the estimated prediction variance

• a data frame containing the original data with the following variables appended:

  1. x-coordinates in UTM

  2. y-coordinates in UTM

  3. sitewise count predictions

  4. sitewise prediction variances

  5. indicator variable for whether or not the each site was sampled

  6. estimated mean for each site

  7. estimated detection probability for each site

  8. the area of each site

• vector with estimated covariance parameters

• the model formula used in slmfit

• the correlation model used

• the prediction weights.

Examples

slmfitobj <- slmfit(formula = Moose ~ CountPred + Stratum,
data = vignettecount,
xcoordcol = "Xcoords", ycoordcol = "Ycoords")
predict(object = slmfitobj)

highamm/FPBK-with-Detection documentation built on Jan. 2, 2022, 6:35 a.m.