BuyseTest_Gehan_cpp and BuyseTest_PetoEfron_cpp functions calls for each endpoint and each strata the pairwise comparison function suited to the type of endpoint and store the resuts.
GPC_cpp(Treatment, Control, threshold, survEndpoint, delta_Treatment,
delta_Control, D, returnIndex, strataT, strataC, n_strata, n_TTE, Wscheme,
index_survivalM1, threshold_TTEM1, list_survivalT, list_survivalC, methodTTE,
correctionTTE, neutralAsUninf, keepComparison)A matrix containing the values of each endpoint (in columns) for the treatment group (in rows). const arma::mat&.
A matrix containing the values of each endpoint (in columns) for the control group (in rows). const arma::mat&.
Store the thresholds associated to each endpoint. const NumericVector&. Must have length D. The threshold is ignored for binary endpoints. Must have D columns.
Does each endpoint is a time to event. const LogicalVector&. Must have length D.
A matrix containing in the type of event (0 censoring, 1 event) for each TTE endpoint (in columns) and treatment observations (in rows). const arma::mat& containing binary integers. Must have n_TTE columns. Ignored if n_TTE equals 0.
A matrix containing the nature of observations in the control group (in rows) (0 censoring, 1 event) for each TTE endpoint (in columns) . const arma::mat& containing binary integers. Must have n_TTE columns. Ignored if n_TTE equals 0.
The number of endpoints. Strictly positive const int.
Should the indexes of the neutral or uninformative pairs be returned. const bool.
A list containing the indexes of treatment observations belonging for each strata. List& of vector containing positive integers.
A list containing the indexes of control observations belonging for each strata. List& of vector containing positive integers.
The number of strata . Strictly positive const int.
The number of time-to-event endpoints. Positive const int.
The matrix describing the weighting strategy. For each endpoint (except the first) in column, weights of each pair are initialized at 1 and multiplied by the weight of the endpoints in rows where there is a 1. const arma::mat&. Must have n_TTE lines and D-1 columns.
The position, among all the survival endpoints, of the last same endpoint (computed with a different threshold). If it is the first time that the TTE endpoint is used it is set to -1. const IntegerVector. Must have length n_TTE.
The previous latest threshold of each TTE endpoint. When it is the first time that the TTE endpoint is used it is set to -1. const NumericVector. Must have length n_TTE.
A list of matrix containing the survival estimates (-threshold, 0, +threshold ...) for each event of the treatment group (in rows). List&. Must have length n_TTE. Each matrix must have 3 (if method is Peto, only one survival function is computed) or 11 (if method is Efron or Peron, two survival functions are computed) columns. Ignored if method is Gehan.
A list of matrix containing the survival estimates (-threshold, 0, +threshold ...) for each event of the control group (in rows). List&. Must have length n_TTE. Each matrix must have 3 (if method is Peto) or 11 (if method is Efron or Peron) columns. Ignored if method is Gehan.
The type of method used to compare censored pairs (1 Peto, 2 Efron, 3 Peron).
Should the uninformative weight be re-distributed to favorable and unfavorable?
Should paired classified as neutral be re-analyzed using endpoints of lower priority? logical.
Should the result of each pairwise comparison be kept? logical.