ABSTRACT
From the study, the new technique for developing probability distributions showed excellent performance, and yielded new distributions at different parameter supports. Consequently, the proposed classical probability distributions exhibit better precision, as they outperformed the existing classical probability distributions. The probability models are identified as Type 1, 2 and 3 distributions; which exhibit asymmetric, symmetric and asymmetric features 97 respectively. Numerically, Type 1 2 and 3 distributions satisfy the probability conditions. Some of the properties derived are the cumulative distribution function, symmetricity, moments and related measures, generating functions, skewness and kurtosis, order statistics, survival, hazard functions, cumulative hazard function, entropy measures, parameter estimation, convolution and inverse cumulative function. Finally, the flexibility of the proposed distributions can be proved by their tractability, robustness, asymptotic stability of the estimator parameters and their ability to show better fit over some known classical and generalized distributions, in data modeling.
