Modeling the Dengue Effective Reproductive Number in the Caraga Region, Philippines Using the Mixture of Generalized Gamma and Lomax (MiGGLo) Distribution

Modeling the effective reproductive number (R eff ) of dengue transmission is inherently challenging due to the stochastic and heterogeneous nature of epidemic dynamics.This study introduces the Mixture of Generalized Gamma and Lomax (MiGGLo) distribution as a novel probabilistic model capable of capturing both the skewed central tendency and heavy-tailed behavior observed in epidemiological data.Theoretical properties, including the kth raw moment, were derived to establish its analytical tractability.A Monte Carlo simulation assessed the performance of Maximum Likelihood Estimation (MLE), Ordinary Least Squares (OLS), and Partial Expectation-Maximization (EM) methods across varying sample sizes.Results revealed that MLE consistently achieved the lowest mean squared error and bias, affirming its superior parameter recovery.Application to dengue R eff data in the Caraga Region, Philippines, demonstrated excellent model fit, with the Generalized Gamma component (αg = 0.7609, θg = 1.2119, τ = 4.1938) representing endemic transmission, and the Lomax component (α l = 19.328,θ l = 10.0, p = 0.05) capturing sporadic outbreak surges.The MiGGLo model effectively characterizes both the central and extreme behaviors of dengue transmission, offering a robust and flexible framework for epidemiological modeling and disease forecasting.