Özet:

We define two new probability distributions, unibimodal symmetric model with proportional hazard function to the normal distribution and unibimodal asymmetric model with proportional hazard function to the skew normal distribution. These models allow adjust censored data with bimodal behavior and high (or low) levels of kurtosis compared with kurtosis of the normal distribution and high (or low) levels of asymmetry. The model parameters are estimated by maximum likelihood and the observed information matrix is determined. The flexibility of the new distribution is illustrated by adjusting a set of real data, the number of molecules of HIV-1 RNA per milliliter of blood measured in individuals with confirmed test of the presence of HIV.

Anahtar Kelimeler: