On the Choice of Linear Regression Algorithms for Biological and Ecological Applications
Journal Title: Annual Research & Review in Biology - Year 2016, Vol 10, Issue 3
Abstract
Model II regression (i.e. minimizing residuals obliquely) is the adequate alternative to Model I regression by Ordinary Least Squares (i.e. minimizing residuals vertically) given the absence of well-established dependence relationships or x measured with error. Yet, it has no perfect solution. Determining the true slope from errors-in-the-variables models requires the errors in x and y estimated from higher order moments. However, their accurate estimation requires enormous data sets and thus they are not applicable to most ecological problems. The alternative Reduced Major Axis (RMA) is dependent on a strict set of assumptions, hardly met with real data, making it prone to bias, whereas Principal Components Analysis (PCA) becomes less reliable with decreasing correlations while x and y presenting approximate variances. We used artificial data (allowing for the determination of the true slope) to demonstrate when RMA or PCA should be preferred. Consequently, we propose using PCA whenever r2+s2x/s2y is higher than 1.5. Otherwise, we suggest generating artificial data manipulated to match the structure of the original, and to test which method provides closer estimates to the input true slope. We provide a user-friendly script to perform this task. We tested the use of RMA and PCA with real data about intraspecific and interspecific biomass-density relations in algae and seagrass, algae frond growth, crustacean and bird morphometry, sardine fisheries and social sciences data, commonly finding widely divergent slope estimates leading to severely biased parameter estimations and model applications. Their analyses support the suggested approach for method selection summarized above.
Authors and Affiliations
Vasco M. N. C. S. Vieira, Joel Creed, Ricardo A. Scrosati, Anabela Santos, Georg Dutschke, Francisco Leitão, Aschwin H. Engelen, Oscar R. Huanel, Marie-Laure Guillemin, Marcos Mateus, Ramiro Neves
Keywords
Related Articles
- EP ID EP351071
- DOI 10.9734/ARRB/2016/25219
- Views 103
- Downloads 0
Nutritional Status of Primary School Children in Kawo District of Kaduna Metropolis, Nigeria
Aims: This study was designed to assess the nutritional status of school children aged 5-12 years in Kawo District, Kaduna State, Nigeria in relation to gender and school background. Study Design: One hundred and forty...
Taxonomy of Earthworm Species Collected from East Coast of Guyana
Earthworms contribute significantly towards the proper functioning of our ecosystem and are environmentally and economically beneficial to the well-being of other organisms. These soil dwelling organisms serve to enhance...
Biochemical Assessment of Picralima nitida Seeds on Oxidative Stress Parameters of Albino Rats
The effects of different percentage concentrations (5%, 10%, 20% and 30%) of ground seeds of Picralima nitida on the concentrations of ascorbic acid (AA), glutathione (GSH) and the activities of superoxide dismutase (SOD...
Antimicrobial Activity of Acacia nilotica ssp. nilotica against Some Causative Agents of Urogenital Infections
Aim: To assess and compare the antimicrobial efficacy of methanol and aqueous extracts derived from Acacia nilotica ssp. nilotica against standard strains of microorganisms and clinical isolates. Study Design: An experi...
Effects of Eruca sativa Leaves Extracts on Testes, Fertility Potential and Testosterone Concentration in Male Rats
Background: Eruca sativa (E.S.) is a worldwide herbaceous plant usually used for salad preparations and in traditional medicine for their therapeutic properties. Aims: This study aimed to determine the effect of ethanol...