Comparative Study of Performance of Neural Networks with Other Non-Parametric Regression Estimators
Abstract: Neural Neural networks have drawn attention to researchers in recent years. This is because they show superiority as a modeling technique for datasets showing nonlinear relationships and thus for both data fitting and prediction abilities. In this study we derive a neural network estimator of finite population mean. This study shows that the mean square error values of the neural network estimator are minimal compared to those of other nonparametric estimators. This implies that neural networks are a better estimation technique for estimating population mean.
Keywords: Neural networks, nonlinear model, nonparametric regression, auxiliary information, survey sampling
How to Cite?: Robert Kasisi, "Comparative Study of Performance of Neural Networks with Other Non-Parametric Regression Estimators", Volume 3 Issue 6, June 2014, International Journal of Science and Research (IJSR), Pages: 1291-1294, https://www.ijsr.net/getabstract.php?paperid=20131418, DOI: https://dx.doi.org/10.21275/20131418