Integration Point: Analysis - Binary Outcome

Gabriel Potvin

October 31, 2025

$\leftarrow$ Go back to the Integration Point: Analysis page

Input Variables

When creating a custom R script, you can optionally use specific variables provided by East Horizon’s engine itself. These variables are automatically available and do not need to be set by the user, except for the UserParam variable. Refer to the table below for the variables that are available for this integration point, outcome, and study objective.

Variable	Type	Description
SimData	Data Frame	Subject data generated in current simulation, one row per subject. To access these variables in your R code, use the syntax: SimData$NameOfTheVariable, replacing NameOfTheVariable with the appropriate variable name. See below for more information.
DesignParam	List	Input parameters which may be needed to compute test statistics and perform tests. To access these variables in your R code, use the syntax: DesignParam$NameOfTheVariable, replacing NameOfTheVariable with the appropriate variable name. See below for more information.
LookInfo	List	Input parameters related to multiple looks. Empty when Statistical Design = Fixed Sample, but still mandatory in the functions CyneRgy::GetDecisionString and CyneRgy::GetDecision. See below for more information.
AdaptInfo	List	Input parameters related to sample size re-estimation. See below for more information. Only applicable when Statistical Design = Group Sequential with Sample Size Re-Estimation.
UserParam	List	Contains all user-defined parameters specified in the East Horizon interface (refer to the Instructions section). To access these parameters in your R code, use the syntax: UserParam$NameOfTheVariable, replacing NameOfTheVariable with the appropriate parameter name.

Variables of SimData

Click here to explore the variables of SimData.

Variables of DesignParam

Click here to explore the variables of DesignParam.

Variables of LookInfo

Click here to explore the variables of LookInfo.

Variables of AdaptInfo

Click here to explore the variables of AdaptInfo.

Expected Output Variable

East Horizon expects an output of a specific type. Refer to the table below for the expected output for this integration point:

Type	Description
List	A named list containing ErrorCode and one or multiple of the following members: Decision, AnalysisTime, TestStat, Delta, CtrlCompleters, TrmtCompleters, CtrlPi, StdError. See below for more information.

The output list can take different forms.

For Study Objective = Two Arm Confirmatory

Option 1 (Decision): Expected Members of the Output List

Members	Type	Description
Decision	Integer	Boundary crossing decision: – 0: No boundary crossed. – 1: Lower efficacy boundary crossed. – 2: Upper efficacy boundary crossed. – 3: Futility boundary crossed (only applicable when Statistical Design = Group Sequential). – 4: Equivalence boundary crossed (not available in East Horizon Explore). You can use the functions CyneRgy::GetDecisionString and CyneRgy::GetDecision to get the decision value. See the template below for the correct usage.
AnalysisTime	Numeric	Optional. Estimate of analysis time. For interim analyses, equivalent to look time; for final analysis, equivalent to study duration.
ErrorCode	Integer	Optional. Can be used to handle errors in your script: – 0: No error. – Positive Integer: Nonfatal error, the current simulation will be aborted, but the next simulation will proceed. – Negative Integer: Fatal error, no further simulations will be attempted.

Note for Statistical Design = Group Sequential: When there is no efficacy boundary to be crossed, the return code of 0 stands for futility in the final look. Similarly, when there is no futility boundary to be crossed, the return code of 0 stands for efficacy in the final look. Use the functions CyneRgy::GetDecisionString and CyneRgy::GetDecision to get decision values in a simple way.

Option 2 (TestStat): Expected Members of the Output List

Members	Type	Description
TestStat	Numeric	Value of appropriate test statistic on Wald ﴾Z﴿ scale.
Delta	Numeric	Estimate of Delta. Required if LookInfo$FutBdryScale = 2 or 3 (futility boundary scale is Delta or conditional power) or Promising Zone Scale = Conditional Power - Estimated. Only applicable when Statistical Design = Group Sequential or Group Sequential with Sample Size Re-Estimation.
CtrlCompleters	Numeric	Number of completers for control arm. Required if LookInfo$FutBdryScale = 3 (futility boundary scale is conditional power) and Conditional power Delta option = Estimated Delta / Estimated HR. Only applicable when Statistical Design = Group Sequential or Group Sequential with Sample Size Re-Estimation.
TrmtCompleters	Numeric	Number of completers for treatment arm. Required if LookInfo$FutBdryScale = 3 (futility boundary scale is conditional power) and Conditional power Delta option = Estimated Delta / Estimated HR. Only applicable when Statistical Design = Group Sequential or Group Sequential with Sample Size Re-Estimation.
CtrlPi	Numeric	Proportion for control arm. Required if LookInfo$FutBdryScale = 3 (futility boundary scale is conditional power) and Conditional power Delta option = Estimated Delta / Estimated HR. Only applicable when Statistical Design = Group Sequential or Group Sequential with Sample Size Re-Estimation.
StdError	Numeric	Standard error of estimate of Delta.
AnalysisTime	Numeric	Optional. Estimate of analysis time. For interim analyses, equivalent to look time; for final analysis, equivalent to study duration.
ErrorCode	Integer	Optional. Can be used to handle errors in your script: – 0: No error. – Positive Integer: Nonfatal error, the current simulation will be aborted, but the next simulation will proceed. – Negative Integer: Fatal error, no further simulations will be attempted.

Note for Statistical Design = Fixed Sample: As the design does not have any futility boundary, TestStat will be used to check for efficacy.

Notes for Statistical Design = Group Sequential:

• If the design does not have any futility boundary, TestStat will be used to check for efficacy.
• If LookInfo$FutBdryScale = 0 (futility boundary scale is Z scale), TestStat will be used to check for both efficacy and futility.
• If LookInfo$FutBdryScale = 2 (futility boundary scale is Delta scale), TestStat will be used to check for efficacy and Delta will be used to check for futility.
• If LookInfo$FutBdryScale = 3 (futility boundary scale is conditional power scale), TestStat will be used to check for efficacy and Delta, CtrlCompleters, TrmtCompleters, and CtrlPi will be used to check for futility.

Notes for Statistical Design = Group Sequential with Sample Size Re-Estimation:

See the notes for Statistical Design = Group Sequential above for futility and efficacy checks. Then, if Promising Zone Scale = Conditional Power - Estimated, Delta and StdError will be used for sample size re-estimation.

For Study Objective = Multiple Arm Confirmatory

Option 1 (Decision): Expected Members of the Output List

Members	Type	Description
Decision	Vector of Integer	Vector of length DesignParam$NumTreatments, containing the boundary crossing decision for each treatment arm: – 0: No boundary crossed. – 1: Lower efficacy boundary crossed. – 2: Upper efficacy boundary crossed. – 3: Futility boundary crossed (only applicable when Statistical Design = Group Sequential). – 4: Equivalence boundary crossed (not available in East Horizon Explore). You can use the functions CyneRgy::GetDecisionString and CyneRgy::GetDecision to get the decision value. See the template below for the correct usage.
AnalysisTime	Numeric	Optional. Estimate of analysis time. For interim analyses, equivalent to look time; for final analysis, equivalent to study duration.
ErrorCode	Integer	Optional. Can be used to handle errors in your script: – 0: No error. – Positive Integer: Nonfatal error, the current simulation will be aborted, but the next simulation will proceed. – Negative Integer: Fatal error, no further simulations will be attempted.

Notes for Statistical Design = Group Sequential:

• Decision should take care of multiple comparison procedures adjustment appropriately.
• When there is no efficacy boundary to be crossed, the return code of 0 stands for futility in the final look. Similarly, when there is no futility boundary to be crossed, the return code of 0 stands for efficacy in the final look. Use the functions CyneRgy::GetDecisionString and CyneRgy::GetDecision to get decision values in a simple way.

Option 2 (TestStat): Expected Members of the Output List

Members	Type	Description
TestStat	Vector of Numeric	Vector of length DesignParam$NumTreatments, containing the value of appropriate test statistic on Wald ﴾Z﴿ scale for each treatment arm.
Delta	Numeric	Estimate of Delta. Required if LookInfo$FutBdryScale = 2 (futility boundary scale is Delta). Only applicable when Statistical Design = Group Sequential.
AnalysisTime	Numeric	Optional. Estimate of analysis time. For interim analyses, equivalent to look time; for final analysis, equivalent to study duration.
ErrorCode	Integer	Optional. Can be used to handle errors in your script: – 0: No error. – Positive Integer: Nonfatal error, the current simulation will be aborted, but the next simulation will proceed. – Negative Integer: Fatal error, no further simulations will be attempted.

Note for Statistical Design = Fixed Sample: As the design does not have any futility boundary, TestStat will be used to check for efficacy.

Notes for Statistical Design = Group Sequential:

• If the design does not have any futility boundary, TestStat will be used to check for efficacy.
• If LookInfo$FutBdryScale = 1 (futility boundary scale is adjusted p-value scale), TestStat will be used to check for both efficacy and futility.
• If LookInfo$FutBdryScale = 2 (futility boundary scale is Delta scale), TestStat will be used to check for efficacy and Delta will be used to check for futility.

Option 3 (AdjPVal): Expected Members of the Output List

Members	Type	Description
AdjPVal	Vector of Numeric	Vector of length DesignParam$NumTreatments, containing the p-values computed from test statistics and adjusted for multiple comparison procedures for each treatment arm.
Delta	Numeric	Estimate of Delta. Required if LookInfo$FutBdryScale = 2 (futility boundary scale is Delta). Only applicable when Statistical Design = Group Sequential.
AnalysisTime	Numeric	Optional. Estimate of analysis time. For interim analyses, equivalent to look time; for final analysis, equivalent to study duration.
ErrorCode	Integer	Optional. Can be used to handle errors in your script: – 0: No error. – Positive Integer: Nonfatal error, the current simulation will be aborted, but the next simulation will proceed. – Negative Integer: Fatal error, no further simulations will be attempted.

Note for Statistical Design = Fixed Sample: As the design does not have any futility boundary, AdjPVal will be used to check for efficacy.

Notes for Statistical Design = Group Sequential:

• If the design does not have any futility boundary, AdjPVal will be used to check for efficacy.
• If LookInfo$FutBdryScale = 1 (futility boundary scale is adjusted p-value scale), AdjPVal will be used to check for both efficacy and futility.
• If LookInfo$FutBdryScale = 2 (futility boundary scale is Delta scale), AdjPVal will be used to check for efficacy and Delta will be used to check for futility.

Option 4 (RawPVal): Expected Members of the Output List

Members	Type	Description
RawPVal	Vector of Numeric	Vector of length DesignParam$NumTreatments, containing the p-values computed from test statistics for each treatment arm.
Delta	Numeric	Estimate of Delta. Required if LookInfo$FutBdryScale = 2 (futility boundary scale is Delta). Only applicable when Statistical Design = Group Sequential.
AnalysisTime	Numeric	Optional. Estimate of analysis time. For interim analyses, equivalent to look time; for final analysis, equivalent to study duration.
ErrorCode	Integer	Optional. Can be used to handle errors in your script: – 0: No error. – Positive Integer: Nonfatal error, the current simulation will be aborted, but the next simulation will proceed. – Negative Integer: Fatal error, no further simulations will be attempted.

Note for Statistical Design = Fixed Sample: As the design does not have any futility boundary, RawPVal will be used to check for efficacy.

Notes for Statistical Design = Group Sequential:

• If the design does not have any futility boundary, RawPVal will be used to check for efficacy.
• If LookInfo$FutBdryScale = 1 (futility boundary scale is adjusted p-value scale), this option cannot be used.
• If LookInfo$FutBdryScale = 2 (futility boundary scale is Delta scale), RawPVal will be used to check for efficacy and Delta will be used to check for futility.

Minimal Templates

Your R script could contain a function such as these ones, with a name of your choice. All input variables must be declared, even if they are not used in the script. We recommend always declaring UserParam as a default NULL value in the function arguments, as this will ensure that the same function will work regardless of whether the user has specified any custom parameters in East Horizon.

Detailed templates with step-by-step explanations are available here: Analyze.Binary.R for 2-Arm and Analyze.Binary.MAMS.R for Multiple Arm.

For Study Objective = Two Arm Confirmatory

Minimal Template for Option 1 (Decision)

For Statistical Design = Fixed Sample

PerformDecision <- function( SimData, DesignParam, LookInfo = NULL, UserParam = NULL )
{
    library( CyneRgy )
    nError              <- 0 # Error handling (no error)
    
    # This is an example using GetDecisionString and GetDecision.
    
    # Write the actual code here.
    # It is a fixed sample design, so no interim look nor futility check.
    bFAEfficacyCheck <- TRUE # If TRUE, declares efficacy.
    # Usually, bFAEfficacyCheck would be a conditional statement such as 'dTValue > dBoundary'.
    
    # These variables are set because it is a fixed sample design.
    nQtyOfLooks          <- 1
    nLookIndex           <- 1 
    nQtyOfPatsInAnalysis <- nrow( SimData )
    nTailType            <- DesignParam$TailType
    
    strDecision <- CyneRgy::GetDecisionString( LookInfo, nLookIndex, nQtyOfLooks,
                                               bFAEfficacyCondition = bFAEfficacyCheck)

    nDecision <- CyneRgy::GetDecision( strDecision, DesignParam, LookInfo )
    
    return( list( Decision = as.integer( nDecision ), ErrorCode = as.integer( nError ) ) )
}

For Statistical Design = Group Sequential

PerformDecision <- function( SimData, DesignParam, LookInfo = NULL, UserParam = NULL )
{
    library( CyneRgy )
    nError              <- 0 # Error handling (no error)
    
    # This is an example using GetDecisionString and GetDecision.
    
    # Write the actual code here.
    # It is a group sequential design, so interim looks and futility check are possible.
    bIAEfficacyCheck <- TRUE # If TRUE, declares efficacy at the interim look.
    bIAFutilityCheck <- FALSE # If TRUE, declares futility at the interim look.
    bFAEfficacyCheck <- TRUE # If TRUE, declares efficacy at the final look.
    # Usually, the Check variables would be conditional statements such as 'dTValue > dBoundary'.
    
    # These variables are from LookInfo because it is a group sequential design.
    nQtyOfLooks          <- LookInfo$NumLooks
    nLookIndex           <- LookInfo$CurrLookIndex
    nQtyOfPatsInAnalysis <- LookInfo$CumCompleters[ nLookIndex ]
    RejType              <- LookInfo$RejType
    TailType             <- DesignParam$TailType
    
    strDecision <- CyneRgy::GetDecisionString( LookInfo, nLookIndex, nQtyOfLooks,
                                               bIAEfficacyCondition = bIAEfficacyCheck,
                                               bIAFutilityCondition = bIAFutilityCheck,
                                               bFAEfficacyCondition = bFAEfficacyCheck )

    nDecision <- CyneRgy::GetDecision( strDecision, DesignParam, LookInfo )
    
    return( list ( Decision = as.integer( nDecision ), ErrorCode = as.integer( nError ) ) )
}

Minimal Template for Option 2 (TestStat)

For Statistical Design = Fixed Sample

ComputeTestStat <- function( SimData, DesignParam, LookInfo = NULL, UserParam = NULL )
{
    nError              <- 0 # Error handling (no error)
    dTestStatistic      <- 0
    
    # Write the actual code here.
    # Store the computed test statistic in dTestStatistic.
    
    return( list( TestStat = as.double( dTestStatistic ), ErrorCode = as.integer( nError ) ) )
}

For Statistical Design = Group Sequential

ComputeTestStat <- function( SimData, DesignParam, LookInfo = NULL, UserParam = NULL )
{
    nError              <- 0 # Error handling (no error)
    dTestStatistic      <- 0
    dDelta              <- 0 # Use if futility boundary scale is Delta or conditional power
    dCtrlCompleters     <- 0 # Use if futility boundary scale is conditional power
    dTrmtCompleters     <- 0 # Use if futility boundary scale is conditional power
    dCtrlPi             <- 0 # Use if futility boundary scale is conditional power
    
    
    # Write the actual code here.
    # Store the computed test statistic in dTestStatistic.
    # Compute dDelta, dCtrlCompleters, dTrmtCompleters, dCtrlPi if needed.
    
    return( list( TestStat = as.double( dTestStatistic ),
                  Delta = as.double( dDelta ), # Include if futility boundary scale is Delta or conditional power
                  CtrlCompleters = as.double( dCtrlCompleters ), # Include if futility boundary scale is conditional power
                  TrmtCompleters = as.double( dTrmtCompleters ), # Include if futility boundary scale is conditional power
                  CtrlPi = as.double( dCtrlPi ), # Include if futility boundary scale is conditional power
                  ErrorCode = as.integer( nError ) ) )
}

For Statistical Design = Group Sequential with Sample Size Re-Estimation

ComputeTestStat <- function( SimData, DesignParam, LookInfo = NULL, UserParam = NULL )
{
    nError              <- 0 # Error handling (no error)
    dTestStatistic      <- 0
    dDelta              <- 0 # Use if futility boundary scale is Delta or conditional power
    dCtrlCompleters     <- 0 # Use if futility boundary scale is conditional power
    dTrmtCompleters     <- 0 # Use if futility boundary scale is conditional power
    dCtrlPi             <- 0 # Use if futility boundary scale is conditional power
    dStdError           <- 0 # Use if Promising Zone Scale = Conditional Power - Estimated
    
    
    # Write the actual code here.
    # Store the computed test statistic in dTestStatistic.
    # Compute dDelta, dCtrlCompleters, dTrmtCompleters, dCtrlPi if needed.
    
    return( list( TestStat = as.double( dTestStatistic ),
                  Delta = as.double( dDelta ), # Include if futility boundary scale is Delta or conditional power
                  CtrlCompleters = as.double( dCtrlCompleters ), # Include if futility boundary scale is conditional power
                  TrmtCompleters = as.double( dTrmtCompleters ), # Include if futility boundary scale is conditional power
                  CtrlPi = as.double( dCtrlPi ), # Include if futility boundary scale is conditional power
                  StdError = as.double(dStdError ), # Include if Promising Zone Scale = Conditional Power - Estimated
                  ErrorCode = as.integer( nError ) ) )
}

For Study Objective = Multiple Arm Confirmatory

Minimal Template for Option 1 (Decision)

For Statistical Design = Fixed Sample

PerformDecision <- function( SimData, DesignParam, LookInfo = NULL, UserParam = NULL )
{
    library( CyneRgy )
    nError              <- 0 # Error handling (no error)
    NumTreatments       <- DesignParam$NumTreatments
    vDecision           <- rep( 0, NumTreatments )  # Initializing decision vector to 0
    
    # This is an example using GetDecisionString and GetDecision.
    # Write the actual code here.
    
    # These variables are set because it is a fixed sample design.
    nQtyOfLooks          <- 1
    nLookIndex           <- 1 
    nQtyOfPatsInAnalysis <- nrow( SimData )
    nTailType            <- DesignParam$TailType
    
    for( i in 1:NumTreatments )
    {
        # It is a fixed sample design, so no interim look nor futility check.
        bFAEfficacyCheck <- TRUE # If TRUE, declares efficacy.
        # Usually, bFAEfficacyCheck would be a conditional statement such as 'dTValue > dBoundary'.
        # This would be different for each treatment arm.
    
        strDecision <- CyneRgy::GetDecisionString( LookInfo, nLookIndex, nQtyOfLooks,
                                                   bFAEfficacyCondition = bFAEfficacyCheck)

        nDecision <- CyneRgy::GetDecision( strDecision, DesignParam, LookInfo )
        vDecision[ i ] = nDecision
    }
    
    
    return( list( Decision = as.integer( vDecision ), ErrorCode = as.integer( nError ) ) )
}

For Statistical Design = Group Sequential

PerformDecision <- function( SimData, DesignParam, LookInfo = NULL, UserParam = NULL )
{
    library( CyneRgy )
    nError              <- 0 # Error handling (no error)
    NumTreatments       <- DesignParam$NumTreatments
    vDecision           <- rep( 0, NumTreatments )  # Initializing decision vector to 0
    
    # This is an example using GetDecisionString and GetDecision.
    # Decision should take care of multiple comparison procedures adjustment appropriately.
    # Write the actual code here.
    
    # These variables are from LookInfo because it is a group sequential design.
    nQtyOfLooks          <- LookInfo$NumLooks
    nLookIndex           <- LookInfo$CurrLookIndex
    nQtyOfPatsInAnalysis <- LookInfo$CumCompleters[ nLookIndex ]
    RejType              <- LookInfo$RejType
    TailType             <- DesignParam$TailType
    
    for( i in 1:NumTreatments )
    {
        # Write the actual code here.
        # It is a group sequential design, so interim looks and futility check are possible.
        bIAEfficacyCheck <- TRUE # If TRUE, declares efficacy at the interim look.
        bIAFutilityCheck <- FALSE # If TRUE, declares futility at the interim look.
        bFAEfficacyCheck <- TRUE # If TRUE, declares efficacy at the final look.
        # Usually, the Check variables would be conditional statements such as 'dTValue > dBoundary'.
        # This would be different for each treatment arm.
    
        strDecision <- CyneRgy::GetDecisionString( LookInfo, nLookIndex, nQtyOfLooks,
                                                   bIAEfficacyCondition = bIAEfficacyCheck,
                                                   bIAFutilityCondition = bIAFutilityCheck,
                                                   bFAEfficacyCondition = bFAEfficacyCheck )

        nDecision <- CyneRgy::GetDecision( strDecision, DesignParam, LookInfo )
        vDecision[ i ] = nDecision
    }
    
    
    return( list( Decision = as.integer( vDecision ), ErrorCode = as.integer( nError ) ) )
}

Minimal Template for Option 2 (TestStat)

For Statistical Design = Fixed Sample

ComputeTestStat <- function( SimData, DesignParam, LookInfo = NULL, UserParam = NULL )
{
    nError              <- 0 # Error handling (no error)
    NumTreatments       <- DesignParam$NumTreatments
    vTestStatistic      <- rep( 0, NumTreatments )  # Initializing test statistic vector to 0   
    
    # Write the actual code here.
    # Store the computed test statistic for each treatment arm in vTestStatistic.
    
    return( list( TestStat = as.double( vTestStatistic ), ErrorCode = as.integer( nError ) ) )
}

For Statistical Design = Group Sequential

ComputeTestStat <- function( SimData, DesignParam, LookInfo = NULL, UserParam = NULL )
{
    nError              <- 0 # Error handling (no error)
    NumTreatments       <- DesignParam$NumTreatments
    vTestStatistic      <- rep( 0, NumTreatments )  # Initializing test statistic vector to 0   
    dDelta              <- 0 # Use if futility boundary scale is Delta
    
    # Write the actual code here.
    # Store the computed test statistic for each treatment arm in vTestStatistic.
    
    return( list( TestStat = as.double( vTestStatistic ),
                  Delta = as.double( dDelta ), # Include if futility boundary scale is Delta
                  ErrorCode = as.integer( nError ) ) )
}

Minimal Template for Option 3 (AdjPVal)

For Statistical Design = Fixed Sample

ComputeTestStat <- function( SimData, DesignParam, LookInfo = NULL, UserParam = NULL )
{
    nError              <- 0 # Error handling (no error)
    NumTreatments       <- DesignParam$NumTreatments
    vAdjPVal            <- rep( 0, NumTreatments )  # Initializing p-value vector to 0  
    
    # Write the actual code here.
    # Store the computed adjusted p-value for each treatment arm in vAdjPVal.
    
    return( list( AdjPVal = as.double( vTestStatistic ), ErrorCode = as.integer( nError ) ) )
}

For Statistical Design = Group Sequential

ComputeTestStat <- function( SimData, DesignParam, LookInfo = NULL, UserParam = NULL )
{
    nError              <- 0 # Error handling (no error)
    NumTreatments       <- DesignParam$NumTreatments
    vAdjPVal            <- rep( 0, NumTreatments )  # Initializing p-value vector to 0  
    dDelta              <- 0 # Use if futility boundary scale is Delta
    
    # Write the actual code here.
    # Store the computed adjusted p-value for each treatment arm in vAdjPVal.
    
    return( list( AdjPVal = as.double( vTestStatistic ),
                  Delta = as.double( dDelta ), # Include if futility boundary scale is Delta
                  ErrorCode = as.integer( nError ) ) )
}

Minimal Template for Option 4 (RawPVal)

For Statistical Design = Fixed Sample

ComputeTestStat <- function( SimData, DesignParam, LookInfo = NULL, UserParam = NULL )
{
    nError              <- 0 # Error handling (no error)
    NumTreatments       <- DesignParam$NumTreatments
    vRawPVal            <- rep( 0, NumTreatments )  # Initializing p-value vector to 0  
    
    # Write the actual code here.
    # Store the computed p-value for each treatment arm in vRawPVal.
    
    return( list( RawPVal = as.double( vTestStatistic ), ErrorCode = as.integer( nError ) ) )
}

For Statistical Design = Group Sequential

ComputeTestStat <- function( SimData, DesignParam, LookInfo = NULL, UserParam = NULL )
{
    nError              <- 0 # Error handling (no error)
    NumTreatments       <- DesignParam$NumTreatments
    vRawPVal            <- rep( 0, NumTreatments )  # Initializing p-value vector to 0
    dDelta              <- 0 # Use if futility boundary scale is Delta
    
    # Write the actual code here.
    # Store the computed p-value for each treatment arm in vRawPVal.
    
    return( list( RawPVal = as.double( vTestStatistic ),
                  Delta = as.double( dDelta ), # Include if futility boundary scale is Delta
                  ErrorCode = as.integer( nError ) ) )
}

Examples

Explore the following examples for more context:

1. 2-Arm, Binary Outcome - Analysis
   • AnalyzeUsingEastManualFormulaNormal.R
   • AnalyzeUsingBetaBinomial.R
   • AnalyzeUsingPropLimitsOfCI.R
   • AnalyzeUsingPropTest.R
2. Multiple Arm, Analysis
   • AnalyzeMultiArmUsingPropTestBonferroni.R