analysis.BPositiveBValueEstimator.p_lilliefors

BPositiveBValueEstimator.p_lilliefors(n=100)

p-value of the Lilliefors test. Weights are not yet implemented. Procedure of the test: 1) Magnitudes are dithered to continuous ones by resampling the distribution of the binned magnitudes, taking into account the exponential distribution of the magnitudes. Here, the b-value estimated from the sample is used. 2) estimate the lilliefors p-value for continuous magnitudes. 3) step 1 and 2 are repeated n times, and the mean p-value is returned. As the p-value is produced by random dithering, the result is also not deterministic, and some fluctuation will occur if the calculation is repeated. Choose a high n in case a higher precision is needed.

Source: - Herrmann, M. and W. Marzocchi (2020). “Inconsistencies and Lurking Pitfalls in the Magnitude-Frequency Distribution of High-Resolution Earthquake Catalogs”. Seismological Research Letters 92(2A). doi: 10.1785/0220200337 - Lilliefors, Hubert W. “On the Kolmogorov-Smirnov test for the exponential distribution with mean unknown.” Journal of the American Statistical Association 64.325 (1969): 387-389.