11.6.2. Defining the TMCMC properties

Once the TMCMC instance is created along with its TDistanceLikelihoodFunction, a few methods are available to tune the algorithms. All of these methods are optional, as default values are provided. However, they should be applied in the following order:

1. Define the MCMC algorithm properties that are shared by all chains: MCMC algorithm, acceptation range, and information display interval (these can be set in any order).

2. Specify the number of chains to initialize.

3. Prepare the starting points and the initial standard deviations of the proposal distribution (these can be set in any order).

• 11.6.2.1. Chosing the MCMC algorithm
• 11.6.2.2. Tuning the acceptation rate
• 11.6.2.3. Information on the process
• 11.6.2.4. Initialising several chains
• 11.6.2.5. Initialising the process