Analysis Stack

AnalysisStack function

The AnalysisStack function can be used to get the results generated in the Clinical Scenario Evaluation. The generation of the analysis stack is based on the pre-defined DataModel, AnalysisModel and SimParameters objects.

Description

Inputs

The AnalysisStack function requires the input of three pre-specified objects defined in the following three arguments:

  • data.model defines a DataModel object, as described in this page.

  • analysis.model defines an AnalysisModel object, as described in this page.

  • sim.parameters defines a SimParameters object, as described in this page.

Outputs

The AnalysisStack function returns an AnalysisStack object containing a list with the following components:

  • description: a description of the object.

  • analysis.set: a list of size n.sims defined in the SimParameters object. This list contains the analysis results generated for each data scenario (first level), and for each test and statistic defined in the AnalysisModel object. The results generated for the ith simulation runs and the jth data scenario are stored in analysis.stack$analysis.set[[i]][[j]]$result (where analysis.stack is a AnalysisStack object). Then, this list is composed of three lists:

    • tests return the unadjusted p-values for to the tests defined in the AnalysisModel object.

    • statistic return the statistic defined in the AnalysisModel object.

    • test.adjust return a list of adjusted p-values according to the multiple testing procedure(s) defined in the AnalysisModel object. The lenght of this list corresponds to the number of MultAdjProc objects defined in the AnalysisModel object. Note that if no MultAdjProc objects have been defined, this list contains the unadjusted p-values.

  • analysis.scenario.grid: a data frame indicating all data and analysis scenarios according to the DataModel and AnalysisModel objects.

  • analysis.structure: a list containing the analysis structure according to the AnalysisModel object.

  • sim.parameters: a list containing the simulation parameters according to SimParameters object.

Example

The following example illustrates the use of the AnalysisStack function.

In this example, we will use the data model and simulation parameters specified in the Case study 1 for normally distributed endpoints (see Case study 1).

Specification of the DataModel object

The data model is specified below:

# Outcome parameter set 1
outcome1.placebo = parameters(mean = 0, sd = 70)
outcome1.treatment = parameters(mean = 40, sd = 70)

# Outcome parameter set 2
outcome2.placebo = parameters(mean = 0, sd = 70)
outcome2.treatment = parameters(mean = 50, sd = 70)

# Data model
case.study1.data.model = DataModel() +
  OutcomeDist(outcome.dist = "NormalDist") +
  SampleSize(c(50, 55, 60, 65, 70)) +
  Sample(id = "Placebo",
         outcome.par = parameters(outcome1.placebo, outcome2.placebo)) +
  Sample(id = "Treatment",
         outcome.par = parameters(outcome1.treatment, outcome2.treatment))

Specification of the AnalysisModel object

The analysis model is specified below:

case.study1.analysis.model = AnalysisModel() +
  Test(id = "Placebo vs treatment",
       samples = samples("Placebo", "Treatment"),
       method = "TTest") +
  Statistic(id = "Mean Treatment",
            method = "MeanStat",
            samples = samples("Treatment"))

Specification of the SimParameters object

Additionally, the simulation parameters are specified:

# Simulation Parameters
case.study1.sim.parameters = SimParameters(n.sims = 1000, proc.load = 2, seed = 42938001)

Data generation

The results generated in the CSE function can be obtained using the AnalysisStack function as follows:

# Generate results
case.study1.analysis.stack = AnalysisStack(data.model = case.study1.data.model,
                                           analysis.model = case.study1.analysis.model,
                                           sim.parameters = case.study1.sim.parameters)

Manipulation of an AnalysisStack object

Once the AnalysisStack object has been generated, the user can manipulate the object easily as it is structure as a list object. Basically, the results sets are stored in the analysis.set level of the object. This length of the list corresponds to the number of simulations specified in the SimParameters object.

# Print all data sets generated for all simulations
case.study1.analysis.stack$analysis.set

Within each simulation run, a nested list is used to store the results set for each data and analysis scenario, for each test and statistic. For example, the user can access to the results generated in the 100th simulation run for the 2nd data scenario as follows:

# Print the data set generated in the 100th simulation run for the 2nd data scenario
case.study1.analysis.stack$analysis.set[[100]][[2]]

ExtractAnalysisStack function

To ease the manipulation of a AnalysisStack object, a function has been created to extract a particular set of results stack according to the data scenario and simulation runs index.

Description

Inputs

The ExtractAnalysisStack function requires the input of a pre-specified object defined in the following two arguments:

  • analysis.stack defines a AnalysisStack object, as described above.

  • data.scenario defines the data scenario index to extract. By default all data scenarios will be extracted.

  • simulation.run defines the simulation run index. By default all simulation runs will be extracted.

Outputs

The ExtractAnalysisStack function returns a list having the same structure as the analysis.set argument of a AnalysisStack object

  • analysis.set: a list of size corresponding to the number of simulation runs specified by the user defined in the simulation.run argument. This list contains the results generated for each data scenario (data.scenario argument) specified by the user.

Example

Returning to our example, the user can extract the results generated for the 100th simulation run and the 2nd data scenario as follows:

case.study1.extracted.analysis.stack =
  ExtractAnalysisStack(analysis.stack = case.study1.analysis.stack,
                       data.scenario = 2,
                       simulation.run = 100)

A carefull attention should be paid on the index of the nested lists. As only one simulation run index has been specified in this example, the result for the 100th simulation runs is stored in the first level of the list (case.study1.extracted.analysis.stack$analysis.set[[1]]). Similarly, as only one data scenario has been requested, the result for the 2nd data scenario is now in the first position (case.study1.extracted.analysis.stack$analysis.set[[1][[1]]).