ISSN: 0041-3216

ISSN: 0041-3216 (Online), 0041-3216 (Print)
Volume 87 Number 1
Research Papers
Application of statistical principles for evolving crop-logging models in banana (cv Ney Poovan). (29)
R. Venugopalan
Venugopalan and Shamsundaran (2005) advocated the use of statistical modeling technique for evolving crop-logging models in Robusta Banana. Along the similar lines, in the present communication statistical models were developed for evolving crop-logging technique with a view to identify biometrical variables which are best indicators of banana (cv Ney Poovan) yield across its various growth stages. Data was collected from farmers field located at Kestur, Bangalore, India during 2003-05 & 2006-08. Efforts were also made to work out point and interval estimates of crop-logging variables with 95% probability interval. Statistical measures were also worked individually for each of the stages to judge the statistical validity of the models. Thus, the stochastic models developed with identically and independently distributed (i.i.d) white noise, revealed that 48 cm of leaf length and 37 cm of plant height at 71 Days after planting (DAP), 64.5 cm of leaf length and 10 leaves at 144 DAP, 90 cm of leaf length and 36 cm of Leaf breadth at 207 DAP, 13 leaves and 41 cm of leaf breadth at 260 DAP, 42.5 cm of leaf breadth and 14 leaves at 326 DAP, 51cm of plant girth at 380 DAP were the best indicators of crop yield. The prediction power of the crop-logging models were worked out be in the range of 93.6% - 96%. Second year results also revealed that 59 cm of leaf length & 25 cm of plant girth at 75 Days after planting (DAP); 74 cm of leaf length & 13 leaves at 150 DAP; 112 cm of leaf length, 36 cm of leaf breadth and 15 leaves at 240 DAP and 52 cm of plant girth and 18 leaves at 370 DAP were the best indicators of crop yield. Before making a final decision about the model adequacy, randomness assumption about the error term was statistically tested.
Keywords: Crop-logging , Ney Poovan, statistical models and white noise