22.1 Introducing Microsimulation using Distributions

In a Microsimulation model, each trial or simulation of the model can have a different set of parameter inputs. When the distributions are referenced in the model, they will have an impact on one or more values within the model.

The method to introduce these different inputs into the model is by assigning distributions to parameters. These distributions are sampled for each parameter before the model is run. TreeAge Pro supports many distribution types. Please refer to the section Creating and referencing distributions for more details.

Open the Legal tutorial example, BreachOfContract-Microsim.trex, and examine the distribution: dist_vConsDamages_1Low.

The distribution,dist_vConsDamages_1Low, provides a range of values for the low consequential damages. These values are sampled from a Gamma distribution with a mean value of $1million and a standard deviation of $50k. Due to the characteristics of the Gamma distribution, you can see the range of samples we would expect in the probability density function graph at the top right within the range (904,391 - 1,100,344).

The sampling rate is set to “Resample per individual trial”, which tells TreeAge Pro to generate a new sampled values for the low consequential damages for every simulation of the model.

It is likely you will want to create distributions to match the parameters you are modelling. For example, the Gamma distribution often provides a good range of values for costs if you have a population that affects men more frequently than women, you will want to assign patient gender based on the distribution of gender in the disease cohort.

Within the model, you can refer to the distribution directly by name. When the variable, in this case value of low consequential damages is needed, the sampled value for that trial will be returned. You can see the Variables and Distributions View below where all the distributions are referenced by the appropriate variable.

Refer to the section Creating and referencing distributions for information on creating and managing distributions.