Modern Statistical Computation and Simulation

Description

You will learn how to use a suite of statistical techniques to bring to bear the wealth of computing power available to your generation. This course will present a variety of statistical techniques which require minimal assumptions, so you can get on with the more useful and some would say more interesting aspects of statistics without worrying about the whether the assumptions underlying more traditional approaches hold for your real life data. It will provide an intuitive and practical understanding of these techniques alongside suitable development of their mathematical underpinnings. An integral part of the lectures and the labs are hands-on components showing you how to implement them in statistical software, to take advantage of the ample computing power available to you in the modern world.

The course will cover simulation of random numbers which can be used to mimic real world phenomena. These simulation ideas will provide a suite of exploratory and inferential techniques to manipulate, visualise and make decisions from complex real world data. In particular, we will cover:
1) random number generators;
2) simulation studies;
3) cross-validation;
4) permutation and resampling methods (in particular bootstrapping)
5) kernel density estimation
To find more information about these topics you can search for these terms on Wikipedia, which demonstrates these flexible and powerful tools.