Discrete Weibull distribution
| Parameters |
scale shape |
|---|---|
| Support | |
| pmf | |
| CDF |
![{\displaystyle \exp \left[-\left({\frac {x}{\alpha }}\right)^{\beta }\right]-\exp \left[-\left({\frac {x+1}{\alpha }}\right)^{\beta }\right]}](../I/m/108e0877183ce48d9a98ec953cbff9c6489ca272.svg)
![{\displaystyle 1-\exp \left[-\left({\frac {x+1}{\alpha }}\right)^{\beta }\right]}](../I/m/70736d03d8cf3850c4325764905bba39bc3c50c3.svg)
In probability theory and statistics, the discrete Weibull distribution is the discrete variant of the Weibull distribution. It was first described by Nakagawa and Osaki in 1975.
Alternative parametrizations
In the original paper by Nakagawa and Osaki they used the parametrization making the cmf with . Setting makes the relationship with the geometric distribution apparent.[1]
Location-scale transformation
The continuous Weibull distribution has a close relationship with the Gumbel distribution which is easy to see when log-transforming the variable. A similar transformation can be made on the discrete-weibull.
Define where (unconventionally) and define parameters and . By replacing in the cmf:
We see that we get a location-scale parametrization:
![{\displaystyle =1-\exp \left[-\left({\frac {x+1}{\alpha }}\right)^{\beta }\right]=1-\exp \left[-\left({\frac {e^{y}}{e^{\mu }}}\right)^{\frac {1}{\sigma }}\right]=1-\exp \left[-\exp \left[{\frac {y-\mu }{\sigma }}\right]\right]}](../I/m/25ccaa7e3641fe8d829b42de25c253a07acfec32.svg)
which in estimation-settings makes a lot of sense. This opens up the possibility of regression with frameworks developed for weibull-regression and extreme-value-theory. [2]
See also
References
- ↑ Nakagawa, Toshio and Osaki, Shunji (1975). "The discrete Weibull distribution". Reliability, IEEE Transactions on. 24: 300–301.
- ↑ Scholz, Fritz (1996). "Maximum Likelihood Estimation for Type I Censored Weibull Data Including Covariates". ISSTECH-96-022, Boeing Information & Support Services. Retrieved 26 April 2016.