Selected article for: "infection time and model parameter"

Author: Neri, Franco M.; Cook, Alex R.; Gibson, Gavin J.; Gottwald, Tim R.; Gilligan, Christopher A.
Title: Bayesian Analysis for Inference of an Emerging Epidemic: Citrus Canker in Urban Landscapes
  • Document date: 2014_4_24
  • ID: 01yc7lzk_36
    Snippet: Goodness-of-fit was tested for parameter estimates from different types of temporal windows using posterior predictive distributions [43] . For each time window (delimited by times t 0 and t 1 , with t 0~0 for cumulative windows), a stochastic, spatially explicit model, based upon Equations 2, with parameter values sampled from the posterior distribution, was used to generate a large number (1000) of replicate epidemics, running from time t 0 (wi.....
    Document: Goodness-of-fit was tested for parameter estimates from different types of temporal windows using posterior predictive distributions [43] . For each time window (delimited by times t 0 and t 1 , with t 0~0 for cumulative windows), a stochastic, spatially explicit model, based upon Equations 2, with parameter values sampled from the posterior distribution, was used to generate a large number (1000) of replicate epidemics, running from time t 0 (with initial conditions set according to the recorded infection status) to time t 1 . Three summary statistics were stored for each simulation: the count of infected trees, I(t), and two spatial statistics, the autocorrelation function C t (d) and the ''time-lagged'' statistic R t t0 (d), described in detail below. The posterior predictive distributions for stored values of I(t), C t (d), and R t t0 (d) (henceforth, simulated summary statistics), at times t corresponding to experimental snapshots, were then compared with the corresponding summary statistics extracted from the experimental dataset (henceforth, experimental summary statistics).

    Search related documents:
    Co phrase search for related documents
    • autocorrelation function and infected tree: 1
    • detail describe and different type: 1
    • detail describe and large number: 1, 2
    • different type and large number: 1, 2, 3, 4, 5, 6, 7, 8
    • different type and parameter value: 1
    • experimental dataset and large number: 1, 2, 3
    • fit goodness and large number: 1, 2
    • fit goodness and parameter value: 1
    • fit goodness and posterior distribution: 1
    • initial condition and large number: 1
    • large number and parameter estimate: 1
    • large number and parameter value: 1, 2, 3, 4
    • large number and posterior distribution: 1, 2
    • parameter estimate and posterior distribution: 1, 2
    • parameter value and posterior distribution: 1