Fuzzy regression interval models for forewarning onion thrips

In this study, data have been taken from Directorate of Onion and Garlic Research (DOGR), Pune. The field trials were sown on different dates at fortnightly intervals (15-Jun, 01-Jul, 15-Jul, 01-Aug, 15-Aug, 01-Sep, 15-Sep, 01-Oct, 15-Oct, 01-Nov, 15-Nov, 01-Dec and 15 Dec.) in different seasons at Pune during 2000 to 2009. An attempt has been made to develop fuzzy regression interval models for forewarning time (crop age) of first appearance of onion thrips (Y₁), time (crop age) of peak population of onion thrips (Y₂) and maximum thrips population (Y₃) using weather variables. The developed fuzzy regression models were compared with the weather indices based regression models for various characters; the study reveals that the average widths for linear regression models vis-a-vis their fuzzy counterparts are much higher for all values of fitness criterion (h). Thus, Fuzzy regression methodology is more efficient than linear regression technique.