A common way to model nonnegative time series is to apply a log transformation and then use classical ARMA techniques. We demonstrate using Canadian Fire Weather Index (FWI) data that simulating from such models can lead to unrealistic data scenarios. Minification models provide another approach to nonnegative time series, but they can be too restrictive. We propose a random coefficient version of these processes, whose stationarity properties we study in this paper. This model has more flexibility than the fixed coefficient version of the process, and we demonstrate that simulated data from this model can be more realistic, and is so for the FWI series.

中文翻译：

---

随机系数最小化过程

对非负时间序列建模的一种常用方法是应用对数变换，然后使用经典的 ARMA 技术。我们演示了使用加拿大火灾天气指数 (FWI) 数据，从此类模型进行模拟可能会导致不切实际的数据场景。缩小模型为非负时间序列提供了另一种方法，但它们可能过于严格。我们提出了这些过程的随机系数版本，我们在本文中研究了其平稳性。该模型比过程的固定系数版本具有更大的灵活性，我们证明了来自该模型的模拟数据可以更真实，FWI 系列也是如此。