| compare {povmap} | R Documentation |
Compare Function
Description
Function compare is a generic function used to assess the quality of
the model-based estimates by comparing them with the direct estimates.
Method compare.fh assesses the quality of the model-based estimates
of the Fay-Herriot model by comparing them with the direct estimates based
on a goodness-of-fit test proposed by Brown et al. (2001) and by
computing the correlation between the regression-synthetic part of the
Fay-Herriot model and the direct estimates.
Usage
compare(object, ...)
## S3 method for class 'fh'
compare(object, ...)
Arguments
object |
an object of type "fh". |
... |
further arguments passed to or from other methods. |
Value
The return of compare depends on the class of its argument.
The documentation of particular methods gives detailed information about the
return of that method.
For the method for class "fh", the null hypothesis, the value W of the test statistic, the degrees of freedom and the p value of the Brown test; and the correlation coefficient of the synthetic part and the direct estimator (Chandra et al. 2015) are returned.
References
Brown, G., R. Chambers, P. Heady, and D. Heasman (2001). Evaluation of small
area estimation methods: An application to unemployment estimates from the UK
LFS. Symposium 2001 - Achieving Data Quality in a Statistical Agency: A
Methodological Perspective, Statistics Canada.
Chandra, H., Salvati, N. and Chambers, R. (2015), A Spatially
Nonstationary Fay-Herriot Model for Small Area Estimation, Journal
of the Survey Statistics and Methodology, 3, 109-135.