Kalman filter for statistical monitoring of forest cover across sub-continental regions [Symposium]

Ray L. Czaplewski

Treesearch

Kalman filter for statistical monitoring of forest cover across sub-continental regions [Symposium]

Informally Refereed

Authors:

Ray L. Czaplewski

Year:

1991

Type:

Paper

Station:

Rocky Mountain Research Station

Source:

In: 4th Symposium on Biometrical Problems in Agriculture, Forestry and Animal Investigations; 19-21 August, Jokioinen, Finland. Washington, DC: The International Biometric Society. 7 p.

Download (PDF 302 KB)

Download (PDF 302 KB): https://research.fs.usda.gov/download/treesearch/33432.pdf

Abstract

The Kalman filter is a generalization of the composite estimator. The univariate composite estimate combines 2 prior estimates of population parameter with a weighted average where the scalar weight is inversely proportional to the variances. The composite estimator is a minimum variance estimator that requires no distributional assumptions other than estimates of the first 2 moments. The Kalman filter recursively combines 2 estimates: a past estimate that is updated for expected change over time, and a current, direct estimate.

Keywords

Kalman filter, forest cover, statistical monitoring, estimate

Citation

Czaplewski, Raymond L. 1991. Kalman filter for statistical monitoring of forest cover across sub-continental regions [Symposium]. In: 4th Symposium on Biometrical Problems in Agriculture, Forestry and Animal Investigations; 19-21 August, Jokioinen, Finland. Washington, DC: The International Biometric Society. 7 p.