Provider: Ingenta Connect
Database: Ingenta Connect
Content: application/x-research-info-systems
TY - ABST
AU - Lappi, Juha
AU - Bailey, Robert L.
TI - Estimation of the Diameter Increment Function or Other Tree Relations Using Angle-Count Samples
JO - Forest Science
PY - 1987-09-01T00:00:00///
VL - 33
IS - 3
SP - 725
EP - 739
KW - log-normal distribution
KW - Increment cores
KW - weighted estimate
KW - mixture distribution
N2 - A formula is derived for the bias when an angle-count sample is used to estimate the mean of a tree variable that is correlated with the breast height diameter. This bias occurs, for instance, if average increment is estimated with increment cores from an angle-count sample. Estimation of mean increment for a given initial diameter is studied further by assuming that increments are log-normally distributed, in which case the sampling distribution is a mixture of three log-normal distributions. An estimate obtained by weighting observations inversely to the basal area (i.e., with the estimated tree frequency) compares favorably in simulations with a parametric estimate derived from the sampling distribution of diameters. If increments are regressed on the initial diameters, then weighting proportionally to the initial basal area and inversely to the current basal area gives smaller bias and standard deviation of parameter estimates than weighting inversely to the current basal area alone. For. Sci. 33(3):725-739.
UR - http://www.ingentaconnect.com/content/saf/fs/1987/00000033/00000003/art00012
ER -