Selected article for: "confidence interval and intervention group"

Author: Eamon B. O’Dea; Harry Snelson; Shweta Bansal
Title: Using heterogeneity in the population structure of U.S. swine farms to compare transmission models for porcine epidemic diarrhoea
  • Document date: 2015_3_27
  • ID: 1xxrnpg3_24
    Snippet: We assumed that the random effect Z i is normally distributed. Then the likelihood is fully specified. We calculated marginal likelihoods with the Laplace approximation and numerically found the parameters that maximised it. In some cases we fixed η to 0.5, which allowed the model to be fully fit with both the lme4 35 and glmmADMB 36 packages in R. 37 To make sure our results were not sensitive to η = 0.5, we used R's optimise function to find .....
    Document: We assumed that the random effect Z i is normally distributed. Then the likelihood is fully specified. We calculated marginal likelihoods with the Laplace approximation and numerically found the parameters that maximised it. In some cases we fixed η to 0.5, which allowed the model to be fully fit with both the lme4 35 and glmmADMB 36 packages in R. 37 To make sure our results were not sensitive to η = 0.5, we used R's optimise function to find the value of η in [0, 5] with highest likelihood. We performed several diagnostic checks of our fits, including checking for signs of nonlinearity with partial residual plots and for signs of temporal autocorrelation in the residuals.

    Search related documents:
    Co phrase search for related documents
    • Try single phrases listed below for: 1