Survival Analysis With Generalized Additive Models : Part III (the baseline hazard) | Statistical Reflections of a Medical Doctor

]]>you need to install the MASS R-package.

Place on top of your R-markdown or R-script the following:

require(MASS)

rgds ]]>

> bf<-bayesfit(lmfit,10000)

when I input that onto R, it gives me an error:

Error in FUN(X[[i]], …) : could not find function "mvrnorm"

thanks

]]>Similar concepts are actually used in frequentist statistics when one calculates standard error and p values etc. In essence one integrates(“marginalizes”) over a large parameter space all other random quantities to summarize the uncertainty in one variable at time.

The major reason you do not see this in nonBayesian statisitics is because there everyone assumes that one is dealing with Gaussian variables in which the marginal is also a Gaussian (normal) variable ]]>

Linda.

]]>