Skip to contents

Sensitivity Analysis

sensitivity_analysis.Rmd

Consider a population with the following parameters:

  • Two sub-populations sized 80000 (A) and 20000 (B)

  • At time zero 20% of both A and B are vaccinated

  • R-Naught is 1.75, infected individuals take a week to recover.

  • With a contact ratio of 2, Group B makes twice as many contacts as group A.

  • 75% of Group A’s contacts are within Group A. 50% of Group B’s contacts are within Group B

  • Susceptibility ratio varies from 0.5 to 2. At 0.5 Group A is twice as susceptible to infection as Group B. At 2, Group A is half as susceptible to infection as Group A.

index <- vector()
groupA <- vector()
groupB <- vector()
popA <- 80000
popB <- 20000

# For each
for (i in seq(0.5, 2, 0.01)) {

  ret <- getFinalSize(vacTime = 0,
    vacPortion = c(0.2, 0.2),
    popSize = c(popA, popB),
    R0 = 1.75,
    recoveryRate = 1 / 7,
    relContact = c(1, 2),
    contactWithinGroup = c(0.75, 0.5),
    relSusc = c(1, i))

  if (ret[1] > 1000) {
    index <- c(index, i)
    groupA <- c(groupA, ret[1] / popA)
    groupB <- c(groupB, ret[2] / popB)
  }


}

print(max(max(groupA), max(groupB)))
#> [1] 0.4683875
plot(index, groupA, type = "l", xlab = "Group B / Group A Susceptibility Ratio", ylab = "Infection Prevalence", ylim = c(0, max(max(groupA), max(groupB))))
lines(index, groupB, type = "l", col = "red")
legend("topright", legend = c("Group A Prevalence", "Group B Prevalence"), col = c("black", "red"), lty = 1)