Title: Improvement of beehive management practices in biology using kernel-based subset vector autoregressive modelling

Authors: Jack Penm

Addresses: School of Finance and Applied Statistics, The Australian National University, Canberra, Australia

Abstract: Honey production in Australia is worth about $50 million a year, with the value of pollination being estimated at between $600 million and $1.2 billion a year. Other bee products such as wax and queen rearing earn about $65 million a year, much of that in export earnings. With reference to current circumstances of reduced production because of the prolonged drought, examination of this industry is crucial as it appears that Australia may be able to increase profitability by improving beehive management practices. In this paper we propose a kernel-based vector recursive algorithm to sequentially estimate subset vector autoregressive models (including full-order models). We apply this algorithm to test the direct causal relationships between the population of honeybee foragers and foraging types gathering nectar, pollen or water. The findings suggest that we may be able to predict the optimal conditions at any time to maximise the honey yield of colonies.

Keywords: beehive management; Monte Carlo integration; subset vector autoregressive modelling; VAR; kernel-based estimation; honey production; Australia.

DOI: 10.1504/IJSTM.2007.013926

International Journal of Services Technology and Management, 2007 Vol.8 No.4/5, pp.389 - 402

Published online: 04 Jun 2007 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article