In Auburngrads/teachingApps: Apps for Teaching Statistics, R Programming, and Shiny App Development

Computing Values For the Hazard Function in R

---

• No predefined R function exists for $h(t)$, but we can take advantage of the fact that $$h(t)=\frac{f(t)}{S(t)}$$

  • For $\;T \sim Weibull(\beta,\theta)$
    $h(t)=$ dweibull(q, shape, scale)/(1 - pweibull(q, shape, scale))

  • For $\;T \sim Exponential(\lambda)$
    $h(t)=$ dexp(q, rate)/(1 - pexp(q, rate))

  • For $\;T \sim Normal(\mu,\sigma)$
    $h(t)=$ dnorm(q, mean, sd)/(1 - pnorm(q, mean, sd))

  • For $\;T \sim Lognormal(\mu,\sigma)$
    $h(t)=$ dlnorm(q, meanlog,sdlog)/(1 - plnorm(q, meanlog, sdlog))

  • For $T \sim Gamma(\kappa,\beta)$
    $h(t)=$ dgamma(q, shape, scale)/(1-pgamma(q, shape, scale))

Auburngrads/teachingApps documentation built on June 17, 2020, 4:57 a.m.