Reading: Novel interval multiple linear regression model to assess the risk of invasive alien plant s...

Download

A- A+
Alt. Display

Articles

Novel interval multiple linear regression model to assess the risk of invasive alien plant species

Authors:

H.O.W. Peiris,

The Open University of Sri Lanka, LK
About H.O.W.
Department of Mathematics, Faculty of Natural Sciences
X close

S. Chakraverty,

National Institute of Technology, RourKela, IN
About S.
Department of MMathematics
X close

S. S. N. Perera ,

University of Colombo, LK
About S. S. N.
Research & Development centre for Mathematical Modelling, Department of Mathematic
X close

S. M. W. Ranwala

University of Colombo, LK
About S. M. W.
Department of Plant Sciences
X close

Abstract

Invasive Alien Species (IAS) can be considered as a serious threat to the existence of the environment as they alter physical, chemical and biological components of the environment. Invasive potential of species can be recognized by their biological traits. Therefore, it is very important to model the risk of species using biological traits before going into a new environment. The purpose of this study is to build interval multiple linear regression models with interval input-output data to evaluate invasion risk of IAS related to biological traits.  A new method has been proposed to estimate the interval regression coefficients. Two different regression models are developed using interval least square algorithm. In the first model we use the method developed by Chenyi Huand the second model is newly developed. The estimated accuracy of the model that is developed by the proposed method is higher in comparison to the model with Chenyi Hu method. These two models are validated using known invasive and non-invasive species. The model that incorporates the proposed method provides better prediction of risk of IAS.

How to Cite: Peiris, H.O.W. et al., (2018). Novel interval multiple linear regression model to assess the risk of invasive alien plant species. Journal of Science. 9(1), pp.12–30. DOI: http://doi.org/10.4038/jsc.v9i1.12
Published on 23 Jul 2018.
Peer Reviewed

Downloads

  • PDF (EN)

    comments powered by Disqus