pomp (version 1.6)

logmeanexp: The log-mean-exp trick

Description

logmeanexp computes $$\log \frac{1}{N}\sum _{n=1}^N{e}_i^x,$$ avoiding over- and under-flow in doing so. It can optionally return an estimate of the standard error in this quantity.

Usage

logmeanexp(x, se = FALSE)

Arguments

x

numeric

se

logical; give approximate standard error?

Value

log(mean(exp(x))) computed so as to avoid over- or underflow. If se = FALSE, the approximate standard error is returned as well.

Details

When se = TRUE, logmeanexp uses a jackknife estimate of the variance in $log(x)$.

Examples

  ## generate a bifurcation diagram for the Ricker map
  pompExample(ricker)
  ll <- replicate(n=5,logLik(pfilter(ricker,Np=1000)))
  ## an estimate of the log likelihood:
  logmeanexp(ll)
  ## with standard error:
  logmeanexp(ll,se=TRUE)