# ROCFTP.MMS

The goal of ROCFTP.MMS is to generate perfect sample from a given posterior. It takes a posterior, and generates two Markov chains by Metropolis with a multishift proposal. This proposal is monotone and forces these two Markov chains started from the two extreme points of the most interest range to be coalesced. This mechanism is used in ROCFTP to generate a perfect sample.

## Installation

You can install the released version of ROCFTP.MMS from CRAN with:

`install.packages("ROCFTP.MMS")`

And the development version from GitHub with:

```
# install.packages("devtools")
devtools::install_github("nabipoor/ROCFTP.MMS")
```

## Example

Basic examples provided in the Help document.