mergeCovmat: Combine sample covariance matrices from two samples

In telefit: Estimation and Prediction for Remote Effects Spatial Process Models

Description

This function combines the sample covariance information from two samples (of the same phenomena) to return the sample covariance matric of the union of the two samples.

Usage

mergeCovmat(
  A.cov.xy,
  B.cov.xy,
  A.mean.x,
  A.mean.y,
  B.mean.x,
  B.mean.y,
  A.n,
  B.n
)

Arguments

A.cov.xy	sample covariance matrix from the first sample, 'A'
B.cov.xy	sample covariance matrix from the second sample, 'B'
A.mean.x	sample mean from the first sample, 'A'
A.mean.y	sample mean from the first sample, 'A'
B.mean.x	sample mean from the second sample, 'B'
B.mean.y	sample mean from the second sample, 'B'
A.n	sample size from the first sample, 'A'
B.n	sample size from the second sample, 'B'

Details

This function assumes the data is normalized by n (the MLE estimator) instead of n-1 (the unbiased estimator).

References

Pebay, P., 2008, Formulas for Robust, One-Pass Parallel Computation of Covariances and Arbitrary-Order Statistical Moments: Sandia Report.

telefit documentation built on Feb. 4, 2020, 9:08 a.m.