Improved multivariate prediction regions for Markov process models

Bootstrap prediction regions for multivariate autoregressive processes

Riassunto: L’obiettivo del presente lavoro è studiare il comportamento di una nuova procedura per la determinazione di regioni di previsione per processi autoregressivi multidimensionali. Le regioni di previsione, basate sulla tecnica bootstrap, non fanno affidamento su alcuna assunzione distributiva per i disturbi ed inoltre tengono conto della variabilità derivante dalla necessità di stimare ...

Multivariate Spatial Process Models

Individual prediction regions for multivariate longitudinal data with small samples.

Follow-up is more and more used in medicine/doping control to identify abnormal results in an individual. Currently, follow-ups are mostly carried out variable by variable using "reference intervals" that contain the values observable in 100(1-α)% of healthy/undoped individuals. Observations of the evolution of the variables over time in a sample of N healthy/undoped individuals, allows these r...

Improved Hidden Markov Models for Speech Recognition

The basic hidden Markov model used in most statistical approaches to speech recognition needs to be considerably expanded and modified in order to become fully efficient. The purpose of this paper is to describe some of these improvements brought to the original scheme. First, good starting values are needed for the training procedure by which the model parameters are adjusted to their optimal ...

Improved estimators for constrained Markov chain models