Modelling counts with state-dependent zero inflation

2018 ◽  
Vol 20 (2) ◽  
pp. 127-147 ◽  
Author(s):  
Tobias A MÖller ◽  
Christian H Weiß ◽  
Hee-Young Kim
Keyword(s):  
Time Series ◽  
The State ◽  
Zero Inflation ◽  
Count Time Series ◽  
Flexible Approach ◽  
Bounded Support ◽  
Excessive Demand ◽  
Parent Distribution ◽  
State Dependent ◽  
Stochastic Properties

We introduce a state-dependent zero-inflation mechanism for count distributions with unbounded or bounded support. Instead of uniformly downweighting the parent distribution, this flexible approach allows us to generate most of the zeros from either low or high counts. We derive the stochastic properties of the inflated distributions and discuss special instances designed for zero inflation caused by, for example, excessive demand or underreporting. Furthermore, we apply the state-dependent zero-inflation mechanism to generalize existing models for count time series with bounded support.

Modeling Zero Inflation in Count Data Time Series with Bounded Support

2017 ◽  
Vol 20 (2) ◽  
pp. 589-609 ◽  
Author(s):  
Tobias A. Möller ◽  
Christian H. Weiß ◽  
Hee-Young Kim ◽  
Andrei Sirchenko
Keyword(s):  
Time Series ◽  
Count Data ◽  
Zero Inflation ◽  
Bounded Support ◽  
Count Data Time Series

scholarly journals Statistical modeling of COVID-19 deaths with excess zero counts

2021 ◽  
Vol 10 (s1) ◽  
Author(s):  
Sami Khedhiri
Keyword(s):  
Time Series ◽  
Negative Binomial ◽  
Forecast Accuracy ◽  
Series Data ◽  
Zero Inflation ◽  
Count Time Series ◽  
Forecast Performance ◽  
Intervention Policy ◽  
Auto Regression ◽  
Zero Counts

Abstract Objectives Modeling and forecasting possible trajectories of COVID-19 infections and deaths using statistical methods is one of the most important topics in present time. However, statistical models use different assumptions and methods and thus yield different results. One issue in monitoring disease progression over time is how to handle excess zeros counts. In this research, we assess the statistical empirical performance of these models in terms of their fit and forecast accuracy of COVID-19 deaths. Methods Two types of models are suggested in the literature to study count time series data. The first type of models is based on Poisson and negative binomial conditional probability distributions to account for data over dispersion and using auto regression to account for dependence of the responses. The second type of models is based on zero-inflated mixed auto regression and also uses exponential family conditional distributions. We study the goodness of fit and forecast accuracy of these count time series models based on autoregressive conditional count distributions with and without zero inflation. Results We illustrate these methods using a recently published online COVID-19 data for Tunisia, which reports daily death counts from March 2020 to February 2021. We perform an empirical analysis and we compare the fit and the forecast performance of these models for death counts in presence of an intervention policy. Our statistical findings show that models that account for zero inflation produce better fit and have more accurate forecast of the pandemic deaths. Conclusions This paper shows that infectious disease data with excess zero counts are better modelled with zero-inflated models. These models yield more accurate predictions of deaths related to the pandemic than the generalized count data models. In addition, our statistical results find that the lift of travel restrictions has a significant impact on the surge of COVID-19 deaths. One plausible explanation of the outperformance of zero-inflated models is that the zero values are related to an intervention policy and therefore they are structural.

scholarly journals Vitamin D, SARS-CoV-2 and Causal Associations in Transversal Studies: The Time-Series Analysis to Reveal Potential Confounders. Comment on Gaudio et al. Vitamin D Levels Are Reduced at the Time of Hospital Admission in Sicilian SARS-CoV-2-Positive Patients. Int. J. Environ. Res. Public Health 2021, 18, 3491

2021 ◽  
Vol 18 (13) ◽  
pp. 6793
Author(s):  
Cristiano Ialongo ◽  
Antonella Farina ◽  
Raffaella Labriola ◽  
Antonio Angeloni ◽  
Emanuela Anastasi
Keyword(s):  
Public Health ◽  
Vitamin D ◽  
Time Series ◽  
Severe Acute Respiratory Syndrome ◽  
Time Series Analysis ◽  
Hospital Admission ◽  
The State ◽  
Series Analysis ◽  
Vitamin D Levels ◽  
Time Of Admission

We read with great interest the paper by Gaudio and colleagues on vitamin D and on the state of patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) at the time of admission [...]

scholarly journals On Uncertainty Relations and Entanglement Detection with Mutually Unbiased Measurements

2015 ◽  
Vol 22 (01) ◽  
pp. 1550005 ◽  
Author(s):  
Alexey E. Rastegin
Keyword(s):  
The State ◽  
Uncertainty Relations ◽  
Trade Off ◽  
Entropic Uncertainty Relations ◽  
Finite Dimensional ◽  
State Dependent ◽  
Entanglement Detection

We formulate some properties of a set of several mutually unbiased measurements. These properties are used for deriving entropic uncertainty relations. Applications of mutually unbiased measurements in entanglement detection are also revisited. First, we estimate from above the sum of the indices of coincidence for several mutually unbiased measurements. Further, we derive entropic uncertainty relations in terms of the Rényi and Tsallis entropies. Both the state-dependent and state-independent formulations are obtained. Using the two sets of local mutually unbiased measurements, a method of entanglement detection in bipartite finite-dimensional systems may be realized. A certain trade-off between a sensitivity of the scheme and its experimental complexity is discussed.

scholarly journals Phase Synchronization Is the Amplified Result by the Hilbert Transform

2015 ◽  
Vol 2015 ◽  
pp. 1-3 ◽  
Author(s):  
Ming-Chi Lu ◽  
Hsing-Chung Ho ◽  
Chen-An Chan ◽  
Chia-Ju Liu ◽  
Jiann-Shing Lih ◽  
...  
Keyword(s):  
Time Series ◽  
Dynamical Systems ◽  
Hilbert Transform ◽  
Phase Synchronization ◽  
Nonlinear Dynamical Systems ◽  
The State ◽  
Nonlinear Dynamical ◽  
Large Correlation ◽  
Current Usage ◽  
The Hilbert Transform

We investigate the interplay between phase synchronization and amplitude synchronization in nonlinear dynamical systems. It is numerically found that phase synchronization intends to be established earlier than amplitude synchronization. Nevertheless, amplitude synchronization (or the state with large correlation between the amplitudes) is crucial for the maintenance of a high correlation between two time series. A breakdown of high correlation in amplitudes will lead to a desynchronization of two time series. It is shown that these unique features are caused essentially by the Hilbert transform. This leads to a deep concern and criticism on the current usage of phase synchronization.

Sign in / Sign up

Export Citation Format

Share Document