Estimation of t0

Estimation of the total \(t_0\) under an almost sure equality between the potential outcomes of each unit. The total can be estimated using the mode selection probabilities or the global answer probabilities.

Usage

estim_tot_0_no_p1(Yobs, modes, pi, p2)

estim_tot_a(Yobs, responses, pi, pa)

Arguments

Yobs

vector of the observed outcomes. (numeric vector of size N the size of the population).

modes

vector of the selected mode of each unit. The first mode of the protocol is defined as "m1", the second as "m2" (character vector or factor of size N).

pi

vector containing the first-order inclusion probabilities (numeric vector of size N).

p2

vector containing the true or estimated selection probabilities of the second mode (numeric vector of size N).

responses

vector indicating which unit answered. If a unit answered then the corresponding value is "r" (character vector or factor of size N).

pa

vector containing the true or estimated global answer probability (numeric vector of size N).

Value

the estimation of the total \(t_0\) (scalar).

Functions

• estim_tot_0_no_p1(): Sum an estimator per mode. Do not use the first mode selection probabilities.

• estim_tot_a(): Use the probability to answer by any mode instead of by each mode.

See also

Other total estimators: estim_tot_m1(), estim_tot_m2(), estim_tot_phi_y()