wdkll_ces: Weighted Double Kernel Local Linear Estimation of Conditional...

In ygeunkim/ceshat: Nonparametric Estimation of Conditional Expected Shortfall

Description

WDKLL estimator of CES

Usage

wdkll_ces(
  formula,
  data,
  prob = 0.95,
  nw_kernel = c("Gaussian", "Epanechinikov", "Tricube", "Boxcar"),
  nw_h,
  pdf_kernel = c("Gaussian", "Epanechinikov", "Tricube", "Boxcar"),
  h0,
  init = 0,
  eps = 1e-05,
  iter = 1000,
  lower_invert = -3,
  upper_invert = 3
)

Arguments

formula	an object class formula.
data	an optional data to be used.
prob	upper tail probability for VaR
nw_kernel	Kernel for weighted nadaraya watson
nw_h	Bandwidth for WNW. If not specified, use the asymptotic optimal.
pdf_kernel	Kernel for initial estimate of conditinal pdf
h0	Bandwidth for pdf kernel. If not specified, use 0.1 times of asymptotic optimal for nw_h.
init	initial value for finding lambda
eps	small value
iter	maximum iteration when finding lambda
lower_invert	lower y when inverting the cdf
upper_invert	upper y when inverting the cdf

Details

Plugging-in in methods gives

\hat{μ}_p(x) = \frac{1}{p} ∑_{t = 1}^n W_{c,t}(x, h) ≤ft[ Y_t \bar{G}_{h_0} (\hat{ν}_p (x) - Y_t) + h_0 G_{1, h_0} (\hat{ν}_p (x) - Y_t) \right]

References

Cai, Z., & Wang, X. (2008). Nonparametric estimation of conditional VaR and expected shortfall. Journal of Econometrics, 147(1), 120-130.

ygeunkim/ceshat documentation built on Dec. 16, 2019, 12:39 p.m.