cqrMMcpp: Function to compute the local composite quantile regression...

In xhuang20/rdcqr: Local Composite Quantile Regression Method for Regression Discontinuity

Description

This function computes the local composite quantile regression estimate. The point of interest can be either interior or boundary.

Usage

cqrMMcpp(x0, x_vec, y, kernID, tau, h, p, maxit = 100L, tol = 1e-04)

Arguments

x0	point of interest
x_vec	a vector of covariates
y	a vector of dependent variable, the treatment outcome variable in the case of regression discontinuity.
kernID	kernel ID for different kernels. kernID = 0: triangular kernel. kernID = 1: biweight kernel. kernID = 2: Epanechnikov kernel. kernID = 3: Gaussian kernel. kernID = 4: tricube kernel. kernID = 5: triweight kernel. kernID = 6: uniform kernel.
tau	A vector of quantile positions. They are obtained by tau = (1:q)/(q+1).
h	A scalar bandwidth.
p	The polynomial degree.
maxit	Maximum iteration number in the MM algorithm for quantile estimation. Defaults to 100.
tol	Convergence criterion in the MM algorithm. Defaults to 1e-4.

Value

cqrMMcpp returns a list with the following components:

beta0	A q by 1 vector of estimates for q quantiles.
beta1	A p by 1 vector of estimates for the first p derivatives of the conditional mean function.

Examples

# Use the Head Start data as an example.
data(headstart)
data_n = subset(headstart, headstart$poverty < 0)
q      = 5
tau    = (1:q) / (q + 1)

# Compute the local composite quantile estimate
est = cqrMMcpp(x0     = 0,
               x_vec  = data_n$poverty,
               y      = data_n$mortality,
               kernID = 2,
               tau    = tau,
               h      = 4.0,
               p      = 1,
               maxit  = 100,
               tol    = 1.0e-4)
               
# Estimate of the conditional mean on the boundary
est_mean = mean(est$beta0)

# Estimate of the first derivative of the conditional mean function
est_d1   = est$beta1[1]

xhuang20/rdcqr documentation built on July 1, 2021, 5:22 a.m.