Fitting Times-to-Failure to a Weibull Distribution
This Demonstration shows the fitting process of times-to-failure (TTF) data to a three-parameter Weibull distribution. The inbuilt function RandomVariate generates a dataset of pseudorandom TTF from a Weibull distribution with "unknown" parameters , , and . While you use the sliders to fit the empirical data to the theoretical distribution, check the goodness-of-fit tests table at the same time, making sure that the -values indicate that there is a significant fit.
During the fitting process you can choose among two graphical displays: either the cumulative distribution function (CDF) or the quantile plot. The "help" option reveals a table with the parameters , , and . The "empirical" parameters (blue) are locked by the "seed" slider and correspond to the TTF data-generating process; the "estimated" parameters (black) correspond to the application of the inbuilt functions EstimatedDistribution and FindDistributionParameters on the TTF dataset.