BOOTSTRAP CONFIDENCE-INTERVALS FOR THE MINIMUM SUM OF ABSOLUTE ERRORS REGRESSION

At present very little is known about inference procedures on the parameters of the minimum sum of absolute errors, MSAE, regression model for small to medium size samples. We propose the use of bootstrap methods for this purpose. The (1 - alpha) confidence intervals on the parameters of the regression model may be constructed by using the bootstrap standard deviation or the bootstrap sampling distribution of the MSAE estimator. We compare and contrast the performance and quality of the intervals obtained by the two methods via a Monte Carlo study.