Abstract
This research a study model of linear regression problem of autocorrelation of random error is spread when a normal distribution as used in linear regression analysis for relationship between variables and through this relationship can predict the value of a variable with the values of other variables, and was comparing methods (method of least squares, method of the average un-weighted, Thiel method and Laplace method) using the mean square error (MSE) boxes and simulation and the study included fore sizes of samples (15, 30, 60, 100). The results showed that the least-squares method is best, applying the fore methods of buckwheat production data and the cultivated area of the provinces of Iraq for years (2010), (2011), (2012), (2013), (2014).
DOI
10.33095/jeas.v21i86.897
Subject Area
Statistical
First Page
675
Last Page
686
Recommended Citation
Ahmed, A. D. (2015). A Comparison Between Some Estimator Methods of Linear Regression Model with Auto-Correlated Errors with Application Data for the Wheat in Iraq. Journal of Economics and Administrative Sciences, 21(86), 675-686. https://doi.org/10.33095/jeas.v21i86.897
