models/full_models/asmaggh.R

In soniamitchell/SpARKjags:

model {
  # Define likelihood model for data:
  for (p in 1:N_patients)
  {
    for(a in 1:antibiotic_classes)
    {
      response[h_GUID[p],a] ~
        dbern(asmaggh.prob[a,
                           sample_month[h_sample_GUID[p]],
                           age_group[h_sample_GUID[p]],
                           gender[h_sample_GUID[p]],
                           hospital[ward[h_sample_GUID[p]]]])
    }
  }

  for (gp in 1:N_gp)
  {
    for(a in 1:antibiotic_classes)
    {
      response[gp_GUID[gp],a] ~
        dbern(asmaggnh.prob[a,
                            sample_month[gp_sample_GUID[gp]],
                            age_group[gp_sample_GUID[gp]],
                            gender[gp_sample_GUID[gp]]])
    }
  }

  for (v in 1:N_volunteers)
  {
    for(a in 1:antibiotic_classes)
    {
      response[v_GUID[v],a] ~
        dbern(asmaggnh.prob[a,
                            sample_month[v_sample_GUID[v]],
                            age_group[v_sample_GUID[v]],
                            gender[v_sample_GUID[v]]])
    }
  }

  for (o in 1:N_outpatients)
  {
    for(a in 1:antibiotic_classes)
    {
      response[o_GUID[o],a] ~
        dbern(asmaggnh.prob[a,
                            sample_month[o_sample_GUID[o]],
                            age_group[o_sample_GUID[o]],
                            gender[o_sample_GUID[o]]])
    }
  }

  # ------------------------

  # Define the priors:
  for(a in 1:antibiotic_classes)
  {
    antibiotic.class.effect[a] ~ dnorm(intercept, tau.class)
    logit(a.prob[a]) <- antibiotic.class.effect[a]
  }

  for(m in 1:N_sample_month)
  {
    samplemonth.effect[m] ~ dnorm(0, tau.samplemonth)
  }

  for (g in 1:N_age_group)
  {
    agegroup.effect[g] ~ dnorm(0, tau.agegroup)
    logit(agegroup.prob[g]) <- agegroup.effect[g]
  }

  gender.effect[female] ~ dnorm(0, 0.001)
  gender.effect[male] <- -gender.effect[female]


  for (h in 1:N_hosp)
  {
    h.effect[h] ~ dnorm(0, tau.h)
    logit(h.prob[h]) <- h.effect[h]
  }

  # equivalent to h.effect
  nh.effect ~ dnorm(0, tau.h)
  logit(nh.prob) <- nh.effect

  for(a in 1:antibiotic_classes)
  {
    for(m in 1:N_sample_month)
    {
      for (r in 1:N_age_group)
      {
        for (g in genders)
        {
          asmagg.effect[a,m,r,g] <- antibiotic.class.effect[a] +
            samplemonth.effect[m] + agegroup.effect[r] + gender.effect[g]

          for (h in 1:N_hosp)
          {
            logit(asmaggh.prob[a,m,r,g,h]) <- asmagg.effect[a,m,r,g] + h.effect[h]
          }
          # Equivalent to asmaggh.prob
          logit(asmaggnh.prob[a,m,r,g]) <- asmagg.effect[a,m,r,g] + nh.effect
        }
      }
    }
  }

  # Prior value for intercept (log-odds of the average resistance in all samples)
  intercept ~ dnorm(0, 0.001)

  # Prior values for precision
  tau.class ~ dgamma(0.001, 0.001)
  tau.samplemonth ~ dgamma(0.001, 0.001)
  tau.agegroup ~ dgamma(0.001, 0.001)
  tau.h ~ dgamma(0.001, 0.001)

  # Convert precisions to sd
  sd.class <- sqrt(1/tau.class)
  sd.samplemonth <- sqrt(1/tau.samplemonth)
  sd.agegroup <- sqrt(1/tau.agegroup)
  sd.hosp <- sqrt(1/tau.h)

  #monitor# full.pd, dic, deviance, a.prob, intercept, sd.agegroup, sd.samplemonth, sd.class, sd.hosp
}