Which of the following is true for the 'Empirical Likelihood' method?
It is a non-parametric method for constructing confidence intervals.
It requires the data to follow a specific parametric distribution.
It allows for the inclusion of side information via constraints on expectations.
It is computationally simpler than the Bootstrap.