Artificial Neural Networks and Time Series of Counts: A Class of Nonlinear INGARCH Models

Article Statistics & Probability

Soft-clipping INGARCH models for time series of bounded counts

Christian H. H. Weiss et al.

Summary: The soft-clipping binomial INGARCH (scBINGARCH) models are proposed as time series models for bounded counts, which have a nearly linear structure and allow for negative autocorrelations. The conditions for the existence and certain mixing properties of the scBINGARCH process are derived, and further stochastic properties are discussed. The consistency and asymptotic normality of maximum likelihood estimators are established, and finite-sample properties are studied with simulations. The practical relevance of the scBINGARCH model's ability to allow for negative parameter and autocorrelation function values is demonstrated with real-data examples.

STATISTICAL MODELLING (2022)

添加到收藏夹

Article Statistics & Probability

SOFTPLUS INGARCH MODELS

Christian H. Weiss et al.

Summary: This paper proposes a novel INGARCH model that uses the softplus function as a response function, which exhibits an approximately linear structure. The model resolves the limitations of existing models in terms of autocorrelation and autoregressive structure. The stochastic properties of the model are derived, and the performance of maximum likelihood estimators for the parameters is analyzed through simulations. The usefulness of the proposed model is demonstrated through three real-data examples.

STATISTICA SINICA (2022)

添加到收藏夹

Article Statistics & Probability