Precision and the probability of misclassification in site index estimation

Abstract

This work presents an estimate of variance for a site index predictor. The methodology was tested on height/age data for loblolly pine from the Coastal Plain of the southeastern United States. We quantified the effects of stand age, base age, and sample size on the variance of site index. The derived variance was used in the development of a measure of the probability of misclassification. We used this measure to show the effects of site index class interval width, stand age, base age, and sample size on classification performancoe f this site index predictor. The results show that changing the width of class intervals or the value of stand age has the greatest effect on the classification performance. The methods have practical value as guides to selecting sample sizes that produce uniformly reliable site index predictions.

Citation

Lloyd, F. Thomas; Hafley, William. 1977. Precision and the probability of misclassification in site index estimation. Forest Science. 23(4): 493-499.