IN SEARCH OF CONVENIENT TECHNIQUES FOR REDUCING BIAS IN THE ESTIMATION OF WEIBULL PARAMETERS FOR UNCENSORED TESTS

Techniques for estimating the parameters of the 2-parameter Weibull di stribution from data obtained from uncensored tests are compared. This will allow the most convenient method to be chosen by considering the data's characteristics and the level of algorithm complexity it is sh own that common techniques, such as least squares regression and maxim um likelihood, may give rise to very significant errors in terms of th e bias of the estimated Weibull parameters. The estimator suggested by Jacquelin for alpha and the correction factor suggested by Ross for b eta usually give errors of < 5 and < 1%, respectively and are to be re commended, because they are straight forward to implement, Techniques are available which will eliminate virtually all bias in the estimatio n of the parameters, but these may be at the expense of considerable c omplexity in implementation.