Table of percentiles for Distribution ID Plot

The percentiles provide the age by which a percentage of the population is expected to fail.

Sometimes it may be difficult to determine the best distribution based on the probability plot and goodness-of-fit measures. Use the table of percentiles to compare the percentiles for several selected distributions to see how your conclusions change depending on the distribution chosen.

If several distributions provide a reasonable fit to the data and similar conclusions, then the choice of the distribution is less important.

However, if your conclusions differ depending on the distribution, you may want to report the most conservative conclusion, collect more data, or use additional information, such as process knowledge and expert advice.

Example output

Table of Percentiles

			Standard Error	95% Normal CI
Distribution	Percent	Percentile	Standard Error	Lower	Upper
Weibull	1	10.0765	2.78453	5.86263	17.3193
Lognormal	1	19.3281	2.83750	14.4953	25.7722
Loglogistic	1	16.0857	2.99498	11.1675	23.1699
Exponential	1	0.809731	0.133119	0.586684	1.11758

Weibull	5	20.3592	3.79130	14.1335	29.3273
Lognormal	5	26.9212	3.02621	21.5978	33.5566
Loglogistic	5	25.5948	3.27675	19.9149	32.8947
Exponential	5	4.13258	0.679391	2.99422	5.70371

Weibull	10	27.7750	4.11994	20.7680	37.1463
Lognormal	10	32.1225	3.09409	26.5962	38.7970
Loglogistic	10	31.5836	3.29780	25.7385	38.7560
Exponential	10	8.48864	1.39552	6.15037	11.7159

Weibull	50	62.6158	4.62515	54.1763	72.3700
Lognormal	50	59.8995	4.31085	52.0192	68.9735
Loglogistic	50	58.6087	4.07051	51.1498	67.1552
Exponential	50	55.8452	9.18089	40.4622	77.0766

Interpretation

Based on a lognormal distribution fitted to the engine windings data, 1% of engine windings are expected to fail by 19.3281 hours.