Inference for Panel Partially Observed Markov Processes
Coercing panelPomp objects as list, pompList or data.frame
Contacts model
Internal function for modifying pparamArray in Mif2
Internal function for modifying pparamArray in Mif2
Get single column or row without dropping names
PIF: Panel iterated filtering
Handling of loglikelihood replicates
Log-mean-exp for panels
#' Create design matrix for panelPomp calculations
Panel Gompertz model
Likelihood for a panel Gompertz model via a Kalman filter
Make a panelPomp model using UK measles data.
Manipulating panelPomp objects
Inference for PanelPOMPs (Panel Partially Observed Markov Processes)
Constructing panelPomp objects
Panel random walk model
Manipulating panelPomp object parameter formats
Particle filtering for panel data
Modifying parameters of filtered objects
panelPomp plotting facilities
Set shared parameters of a panelPomp object
Extract shared parameters from a panelPomp object
Simulations of a panel of partially observed Markov process
Set unit-specific parameters of a panelPomp object
Extract unit-specific parameters from a panelPomp object
Extract units of a panel model
Extract log likelihood of units of panel models
Interpret shortcuts for sQuote()s and dQuote()s in character objec...
Data analysis based on panel partially-observed Markov process (PanelPOMP) models. To implement such models, simulate them and fit them to panel data, 'panelPomp' extends some of the facilities provided for time series data by the 'pomp' package. Implemented methods include filtering (panel particle filtering) and maximum likelihood estimation (Panel Iterated Filtering) as proposed in Breto, Ionides and King (2020) "Panel Data Analysis via Mechanistic Models" <doi:10.1080/01621459.2019.1604367>.