split_bandit: MAB simulation with partitioned context
In dfcorbin/npbanditC: (Contextual) Multi-Armed Bandit Simulations

MAB simulation with partitioned context

1	split_bandit(nsteps, means, sd, k, J, alpha = 1, b = 1)

`nsteps`	The number of steps in the simulation.
`means`	A list of functions representing the true mean of the levers.
`sd`	A vector of the standard deviations of each lever.
`k`	The number of groups to partition the context vector into.
`J`	The number of basis functions to use inside each group.
`alpha`	Parameter to control the rate of decay of the prior variance.
`b`	Hyper parameter for the inverse gamma distribution. Directly controls the level of exploration.