Abstract
© 2017, Springer International Publishing AG. We present a model for under-reported time series count data in which the underlying process satisfy an INAR(1) structure. Parameters are estimated through a naïve method based on the theoretical expression of the autocorrelation function of the underlying process, and also by means of the forward algorithm. The hidden process is reconstructed using the Viterbi algorithm, and a real data example is discussed.
Original language | English |
---|---|
Title of host publication | Trends in Mathematics |
Pages | 29-34 |
Number of pages | 5 |
Volume | 7 |
DOIs | |
Publication status | Published - 1 Jan 2017 |