Forecasting models for the yield of coconut

Loading...
Thumbnail Image
Date
1985
Journal Title
Journal ISSN
Volume Title
Publisher
Department of Statistics, College of Veterinary and Animal Sciences, Mannuthy
Abstract
An analysis of the yield data of 91 coconut palms, maintained at Coconut Research Station (Nileshwar I), Regional Agricultural Research Station, Pilicode, under Kerala Agricultural University and the weather data for the region of Pilicode, collected from Central Plantation Crops Research Institute (CPCRI), Kasaragod District, Kerala was carried out with the following views and objectives. 1) To develop a suitable and reliable statistical methodology for the pre-harvest forecast of coconut crop yields by evolving different empirical-statistical crop- weather models using the original and generated weather variables as predictor variables. 2) To perform a comparative study of relative efficiency, adequacy and performance of each of these crop- forecasting models evolved and to select the 'best1, most promising and plausible crop forecasting models for the purpose of future use in predicting the coconut crop yields reliably in advance of harvest, 3) To investigate the effect and influence of changes in weather variables on the yield of coconut crop, based on the crop forecasting models selected as the 'best' fitted models. 4) To render suggestion and guidelines for further development of statistical crop-weather models, criteria for their selection, and relevant statistical analysis, In this study, the twelve crop forecasting models for the yields of coconut were developed and fitted under the effective crop season of 3 years (i.e., as far back as 36 months from the first month just before a half-year harvest) with 3-month and 6-month period (season), using the generated weather predictor variables. The response variable was taken as average yield of nuts per bearing tree per half year, and the original weather variables were total rainfall, duration of bright sunshine hours, wind velocity, relative humidity and maximum temperature. Since the relative humidity is expressed in percentages, the datawere transformed into arc—sine root proportion.
Description
PG
Keywords
null
Citation
Collections