MV_ise_heter_critical {mlrv}

MV method

Description

Selection of smoothing parameters for bootstrap tests by choosing the index minimizing the volatility of bootstrap statistics or long-run variance estimators in the neighborhood computed before.

Usage

MV_ise_heter_critical(critical, neighbour)

Arguments

critical

a matrix of critical values

neighbour

integer, number of neighbours

Value

a list of results,

• minp: optimal row number

• minq: optimal column number

• min_ise: optimal value

References

Bai, L., & Wu, W. (2024). Difference-based covariance matrix estimation in time series nonparametric regression with application to specification tests. Biometrika, asae013.

Examples

param = list(d = -0.2, heter = 2,
 tvd = 0, tw = 0.8, rate = 0.1,
 cur = 1, center = 0.3, ma_rate =  0,
 cov_tw =  0.2, cov_rate = 0.1,
 cov_center = 0.1, all_tw  = 1, cov_trend = 0.7)
n = 1000
data = Qct_reg(n, param)
p = ncol(data$x)
t = (1:n)/n
B_c = 100 ##small value for testing
Rc = array(rnorm(n*p*B_c),dim = c(p,B_c,n))
result1 = LocLinear(0.2, t, data$y, data$x)
gridm = c(3,4,5)
gridtau = c(0.2, 0.25, 0.3)
critical <- MV_critical(data$y, result1, Rc, gridm, gridtau)
mv_result = MV_ise_heter_critical(critical,  1)
m = gridm[mv_result$minp + 1]
tau_n = gridtau[mv_result$minq + 1]