Zeitsperre
commented
1 year ago Hi,
I just want to let you know that lmoments3 development has switched hands and migrated to https://github.com/Ouranosinc/lmoments3. This repository is essentially obsolete. If you are still having issues or would like to see changes to lmoments3, please feel free to re-open this issue at the new location.
All the best,
I am having problems with the Wakeby distribution in your library.
I am trying to calculate the 0.5 quantile for the following data:
data=[ 67.65056849, 74.38059133, 92.17499793, 70.07513288, 85.01663804, 60.74476708, 57.8690283 , 58.19832142, 45.58640029, 52.37878952, 57.12557528, 52.26964206, 103.2897938 , 147.10877761, 48.85905404, 60.80855671, 114.31812868, 64.95975126, 61.2292042 , 74.81203638, 75.37807133, 48.27653356, 45.17146591, 69.72044762, 51.51300766, 72.7116555 , 62.80461308, 55.12026027, 50.11876859, 81.86401986, 72.79674355, 102.09984854, 53.2433996 , 49.94688258, 49.43813961, 78.94880231, 82.54409563, 77.62494106, 144.03982423, 74.15119205, 56.95598256, 55.86952697, 91.05071314, 56.49325624, 54.72556818, 60.55818591, 71.07502688, 62.97981832, 55.69969788, 95.64060979]
paraEarly = distr.wak.lmom_fit(data) fitted_wak = distr.wak(**paraEarly) UQEarly = fitted_wak.ppf(0.5)
Which return 317.79357036. Clearly not the 50% quantile for the data above. I checked the results with lmoments in python 2 and lmom in RStudio and both libraries give me the right result (63.7252).
Can you please take a look at this or let me know what is the problem.
Thank you,
Luciana