Monte Carlo Vs Bootstrap
After 11,741 days of my existence on this planet, I have understood something very clearly.
The difference between Bootstrapping and Monte Carlo Simulation.
If you know the population distribution parameters, then you can simulate from that distribution and calculate whatever you want. Lets say E(x^3) .. simulate a ton of observations , find out the expectation and you are done with it
Bootstrapping is a little different. Instead of sampling from the population distribution, which we do not know in practice, we sample from the data and calculate whatever metric that we need to compute. One does this a number of times and then you use it come up with whatever you want
to estimate
Such clear understanding I am having after ages…2 lines from Faraway provided me with this insight
I am so happy :)