タイトル | Incorporating partially identified sample segments into acreage estimation procedures: Estimates using only observations from the current year |
本文(外部サイト) | http://hdl.handle.net/2060/19820024921 |
著者(英) | Sielken, R. L., Jr. |
著者所属(英) | Texas A&M Univ. |
発行日 | 1981-12-01 |
言語 | eng |
内容記述 | Several methods of estimating individual crop acreages using a mixture of completely identified and partially identified (generic) segments from a single growing year are derived and discussed. A small Monte Carlo study of eight estimators is presented. The relative empirical behavior of these estimators is discussed as are the effects of segment sample size and amount of partial identification. The principle recommendations are (1) to not exclude, but rather incorporate partially identified sample segments into the estimation procedure, (2) try to avoid having a large percentage (say 80%) of only partially identified segments, in the sample, and (3) use the maximum likelihood estimator although the weighted least squares estimator and least squares ratio estimator both perform almost as well. Sets of spring small grains (North Dakota) data were used. |
NASA分類 | EARTH RESOURCES AND REMOTE SENSING |
レポートNO | 82N32797 NAS 1.26:169169 E82-10375 NASA-CR-169169 FC-T2-04254 TR-21 |
権利 | No Copyright |
|