Authors : S. S. Desai
Page Nos : 232-235
Description :
A popular data analysis technique used in almost all subjects including Biosciences and Agriculture is regression. In regression analysis a researcher has to face with a data on two or more variables and the interest lies in modeling the relationship between them. Mostly, this could be done by using least squares (LS) method or maximum likelihood estimator (MLE) method. Regression model is fitted under certain assumptions like, independence of predictors; error variable follows normal distribution with constant variance etc. A real life data may not satisfy some assumptions and these methods give misleading results. In this article, we use support vector machine for prediction of future values of response variable and the performance of different estimation methods is evaluated through real data.
Keywords: Least squares method; Multiple linear regression; M-estimator; Support vector regression; Prediction risk.
