Local polynomial regression estimation of trawl size selectivity parameters using genetic algorithm


Abstract views: 166 / PDF downloads: 98

Authors

  • C. G. Joshy ICAR- Central Institute of Fisheries Technology
  • N. Balakrishna Department of Statistics, CUSAT, Cochin
  • V. R. Madhu ICAR-CIFT, Cochin

Abstract

This study used a local polynomial generalised linear model to estimate the trawl selectivity curve and its parameters. This modeling technique was applied to trawl selectivity data obtained from the codend selectivity studies of the Dussumier’s anchovy Thryssa dussumieri, an important trawl resource along the Gujarat coast, India, for 40 mm diamond and square mesh codends. The results of this model were compared with the results obtained from the parametric approach and found to have a superior fit based on the model performance statistics. Genetic algorithm was used to estimate the trawl selectivity parameters by minimising the objective function, i.e., estimated squared distance from target. The nonparametric approach was used to estimate the trawl selectivity parameter values (L50 and SR) for two species viz., Upeneus moluccensis and Trichiurus lepturus to confirm its superiority over the parametric approach.

Author Biographies

  • C. G. Joshy, ICAR- Central Institute of Fisheries Technology

    Scientist

    Fish Processing Division

  • N. Balakrishna, Department of Statistics, CUSAT, Cochin

    Professor

    Department of  Statistics, CUSAT, Cochin

  • V. R. Madhu, ICAR-CIFT, Cochin

    Fishing Technology Division, ICAR-CIFT, Cochin

Downloads

Submitted

2018-02-28

Published

2018-09-30

Issue

Section

Articles

How to Cite

Joshy, C. G., Balakrishna, N., & Madhu, V. R. (2018). Local polynomial regression estimation of trawl size selectivity parameters using genetic algorithm. Indian Journal of Fisheries, 65(3). https://epubs.icar.org.in/index.php/IJF/article/view/76544