Abstract
In the frame of the "Rapid Crop Area Estimates in the European Community" we use a sample of squared segments (pieces of land) of 49 ha. each; estimates are made for difference of crop areas between years. The optimum size seems to be larger than the current one, and much larger if ground survey data are obtained by photo-interpretation. The main purpose of this paper is to assess a method, based on splitting the segments into pieces of 1/4 ha., to forecast the precision of estimates with larger segments. The tests made for France confirm the belief that better estimates can be obtained with a fixed cost by enlarging the size of the segments .
We consider as well the problem of assessing smaller segments, for which an easier technique can be used .
Keywords
Areaframe sampling, segment size, intra-cluster correlation
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Recommended Citation
Carfagna, E. and Gallego, F. J.
(1994).
"EXTRAPOLATING INTRA-CLUSTER CORRELATION TO OPTIMIZE THE SIZE OF SEGMENTS IN AN AREA FRAME.,"
Conference on Applied Statistics in Agriculture.
https://doi.org/10.4148/2475-7772.1364
EXTRAPOLATING INTRA-CLUSTER CORRELATION TO OPTIMIZE THE SIZE OF SEGMENTS IN AN AREA FRAME.
In the frame of the "Rapid Crop Area Estimates in the European Community" we use a sample of squared segments (pieces of land) of 49 ha. each; estimates are made for difference of crop areas between years. The optimum size seems to be larger than the current one, and much larger if ground survey data are obtained by photo-interpretation. The main purpose of this paper is to assess a method, based on splitting the segments into pieces of 1/4 ha., to forecast the precision of estimates with larger segments. The tests made for France confirm the belief that better estimates can be obtained with a fixed cost by enlarging the size of the segments .
We consider as well the problem of assessing smaller segments, for which an easier technique can be used .