lbsample: Generates length bias samples from a parametric family.
In lbiassurv: Length-biased correction to survival curve estimation.

Generates length bias samples from a parametric family.

1	lbsample(n, family, par = list(shape, rate, meanlog, sdlog), censor.vec = rexp(n))

`n`	An integer indicates the number of the desired samples.
`family`	A string. Options are `weibull`, `gamma`, `exponential`, `lognormal` , and `loglogistic`.
`par`	A list of parameters related to the chosen `family`, `meanlog`, and `sdlog` for `lognormal`, and `shape` and `rate` for the rest.
`censor.vec`	An integer vector of the censoring indicator. When the generated sample is bigger than the values of the vector, the function return 1 in the output, meaning that sample is right censored.

This function uses the sampling scheme described in Younger(2011).

weibull: f(x)= k*b*(x*b)^(k-1)*exp(1(x*b)^k)

x>=0, rate=b>0, shape=k>0.

exponential: f(x)=lambda*exp(-lambda*x)

x>=0, rate=lambda>0.

lognormal: f(x)=[1/(x*sigma*sqrt(2*pi))]*exp(-[(ln(x)-mu)^2]/2*sigma^2)

x>0, sdlog=sigma>0, meanlog=mu.

loglogistic: f(x)=[(b/a)*(x/a)^(b-1)]/[1+(x/a)^b]^2

x>=0, rate=1/a>0, shape=b>0.

`time`	A vector of `n` random length bias samples from the density function specified in `family`, giving the residual lifetime.
`censor`	A vector of `n` boolean integers to indicates if the sample is censored or not, 1 for censored samples and 0 for uncensored.
`onset`	Starting time of the follow up for each individual, sampled from uniform distribution.