createParDef: Create a parameterDef Object.

In ezsim: provide an easy to use framework to conduct simulation

Description

createParDef creates a new parameterDef object from a list of scalar parameters and a list of other parameters. parameterDef is a short hand of "parameter definition". It defines parameters used by the dgp which is the most important part of a simulation. For each simulation,There is a particular set of parameter. parameterDef allow us to define several parameters for different simulation at once. There are two types of parameter in parameterDef, scalar parameters and other parameters. Scalar parameters must be a scalar. Any vectors or matrix is regarded as a sequence of scalar parameters. For example, n=seq(10,50,10), first simulation takes n=10, second simulation takes n=20 and so on. Other parameters can be anything and it is banker over the scalar parameters. For example, we would like to know how would the sample size affect the variance of the sample mean of normally distributed variable. We can set n=seq(10,50,10), mean=1 and sd=2. (see example)

Usage

createParDef(selection = list(), banker = list())

Arguments

selection	A list of scalar parameters
banker	A list of other parameters

Value

A parameterDef object

Author(s)

TszKin Julian Chan [e-mail]

See Also

setBanker.parameterDef,setSelection.parameterDef,evalFunctionOnParameterDef,generate.parameterDef

Examples

par_def1<-createParDef(selection=list(mean=1,sd=2,n=seq(10,50,10)))

par_def2<-createParDef()
setSelection(par_def2,mean=1,sd=2,n=seq(10,50,10))

identical(par_def1,par_def2)

evalFunctionOnParameterDef(par_def1, function() rnorm(n,mean,sd) )  # 10 random number
evalFunctionOnParameterDef(par_def1, function() rnorm(n,mean,sd), index=3)  # 30 random number

generate(par_def1)

# More than 1 selection parameters
par_def3<-createParDef(selection=list(sd=2,mean=1:3,n=seq(10,50,10)))

generate(par_def3)

ezsim documentation built on May 1, 2019, 8:04 p.m.