We propose a novel class of count time series models alternative to the classic Galton-Watson process with immigration (GWI) and Bernoulli offspring. A new offspring mechanism is developed and its properties are explored. This novel mechanism, called geometric thinning operator, is used to define a class of modified GWI (MGWI) processes, which induces a certain non-linearity to the models. We show that this non-linearity can produce better results in terms of prediction when compared to the linear case commonly considered in the literature. We explore both stationary and non-stationary versions of our MGWI processes. Inference on the model parameters is addressed and the finite-sample behavior of the estimators investigated through Monte Carlo simulations. Two real data sets are analyzed to illustrate the stationary and non-stationary cases and the gain of the non-linearity induced for our method over the existing linear methods. A generalization of the geometric thinning operator and an associated MGWI process are also proposed and motivated for dealing with zero-inflated or zero-deflated count time series data.
翻译:我们建议采用新型的计算时间序列模型,以替代典型的Galton-Watson进程(GWI)和Bernoulli后代的计算时间序列模型。我们开发了新的后代机制并探索其特性。这个称为几何减速操作器的新机制用于界定经修改的GWI进程类别,这给模型带来了某种非线性。我们表明,与文献中通常考虑的线性案例相比,这种非线性在预测方面可以产生更好的结果。我们探讨了我们MGWI进程的固定和非静止版本。对模型参数的推论和通过蒙特卡洛模拟调查的估算员的有限抽样行为进行了探讨。对两套真实数据集进行了分析,以说明固定和非静止案例以及我们方法在现有线性方法上产生的非线性收益。还提议并激励了几何减减薄操作器及相关的MGWI进程,以便处理零膨胀或零减缩计时序列数据。