Let's say that I have a vector $u$ of real values (say of length 100). I know that $u-v$ is normally distributed with mean $\mu$ and variance $\sigma^2$. I would like to calculate a vector of corresponding $v$ values.

So what I'm doing is like this: use $rnorm()$ to sample 10000 random values with mean $\mu$ and variance $\sigma^2$. Then for every value $u_i$ in $u$, I check which quantile $u_i$ is and then pick the value with the same quantile from the randomly generated values:

gen = rnorm(10000, mean=mu, sd=sigma)
for(i in 1:length(u)){
  q = sum( u <= u[i] )/length(u)
  v[i] = u[i] - quantile(gen, probs=q)

Is this correct? Are there any better ways to do it?

Is there a specific term for this procedure?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.