Modelling yield risk measures of major crop plants

The paper deals with the problem of modelling yield risk measures for major crop plants in Poland. Hence, in some cases the gamma distribution offers a better fit to the data than normal distribution, and in addition to linear models, generalized linear models were also used. The research was based on data from Polish FADN, with sample sizes ranging from 416 up to 2300, depending on the crop plant. It was found that models based on the farm level data, can explain on average 20% of variation coefficient unevenness. The most important variables were average yield, type of farming, arable area and land quality. The elimination of the average yield from the models reduced the average determination coefficient to about 9%.

Issue Date:
Feb 23 2012
Publication Type:
Conference Paper/ Presentation
Record Identifier:
PURL Identifier:
Total Pages:
JEL Codes:
Q10; C46

 Record created 2017-04-01, last modified 2018-01-22

Download fulltext

Rate this document:

Rate this document:
(Not yet reviewed)