Data Composition

Simulate variables and data

simulation
distributions

The purpose of this post is to enable readers to create data from scratch which they can use for their analyses or visualizations.

Simulate dependent variables

simulation
correlation

When you simulate a dataset it is often not enough to have independent variables, but you want to have some dependency between the variables. In this post we explore ways of creating this dependency.

Data simulation with rmultinom

simulation
rmultinom

When creating several datasets that depend on each other, the rmultinom function from the stats package can be a useful helper. In this example we will see how to create customer transactions from a customer table.

Useful packages for data simulation

simulation
packages

We will explore the packages wakefield, rcorpora, charlatan, fabricatr, and GenOrd which can be helpful for data simulation.

More articles »

Data Composition