compute_asymptotic_variance_moment_vector: This function computes the asymptotic variance on the...

In cgaillac/MarginalFElogit: Estimation of Average Marginal Effects in Fixed Effect Logit Models

This function computes the asymptotic variance on the estimated moment for a given individual, that results from the (non-parametric) estimation of the P(S = j | X)'s.

Description

This function computes the asymptotic variance on the estimated moment for a given individual, that results from the (non-parametric) estimation of the P(S = j | X)'s.

Usage

compute_asymptotic_variance_moment_vector(
  PSt_X,
  Xdensity,
  Zt_mat,
  nbIndls,
  dimX,
  h_local_lin_proba,
  m,
  c0
)

Arguments

PSt_X	a vector of length Tmax + 1 containing in j-th position the estimate for P(S = j - 1 | X) for the observation at hand.
Xdensity	the estimated density for the covariates (X_1, ..., X_T) for the observation at hand
Zt_mat	a matrix of size (Tmax + 1) x (Tmax + 1) containg in position (i, j) the value of Z_i(X, j, beta) for the observation at hand. Thus, c = Zt_mat * PSt_X and m = (c_1, ..., c_T) / c_0, from which we deduce the asymptotic variance.
nbIndls	the total number of individuals used to estimate the P(S = j | X)'s.
dimX	the total number of covariates at a given period.
h_local_lin_proba	vector of length Tmax + 1 where the j-th element is the bandwidth used to estimate P(S = j - 1 | X).
m	estimated moment vector (length Tmax) for the observation at hand. Starts with the 1st moment.
c0	estimated value of c_0.

Value

a matrix of size Tmax x Tmax, an empirical estimate for the value of the asymptotic variance of the vector of moments m.

cgaillac/MarginalFElogit documentation built on Dec. 24, 2024, 3:23 p.m.