Summary
The simulations shown in this chapter are of two different kinds: model-based simulation and design-based simulation. Model-based simulations simulate data from a certain (super-population) model. We saw that model-based simulations are easy to set-up. The aim is to always know true parameters – here, from the model that simulates random distributions of interest. The estimation is applied to each of the simulated data and compared with the true parameter values.
Design-based simulation studies differ in that sense that the sampling design must be incorporated. This is why we firstly showed how to simulate a finite population from where samples can be drawn. Whenever data sets are sampled with simple random sampling, there is no need for ...
Get Simulation for Data Science with R now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
