Abstract: For testing a linear hypothesis in a censored regression (or censored ``Tobit'') model, three test criteria and four test statistics based on least absolute deviations estimates of parameters are proposed and their limiting chi-square distributions are established. Some consistent estimates of nuisance parameters are obtained for use in computing the test statistics. A simulation study for small sample performance of these test statistics is made by using iterative linear programming.
Key words and phrases: Censored regression model, least absolute deviations method, limiting distribution, linear hypothesis testing.