Determination of the Critical Correlation Coefficient to Establish a Good Fit for Weibull and Log-Normal Failure Distributions 1999-01-0057
Least-squares rank regression is often used for curve fitting Weibull or Normal failure data. Most software packages calculate the correlation coefficient, r, or the coefficient of determination, r2. Monte Carlo simulation was used to create and analyze several thousand data sets. By comparing r from a given sample to results obtained from Monte Carlo simulation, the goodness of fit of the sample can be determined.