demo/nlpr.R

In mpiktas/midasr: Mixed Data Sampling Regression

set.seed(131313)

nnbeta <- function(p, k) nbeta(c(1, p), k)

do_n_lstr <- function(n) {
  dgp <- midas_lstr_sim(n, m = 12, theta = nnbeta(c(2, 4), 24), intercept = 1, plstr = c(1.5, 1, 1, 1), ar.x = 0.9, ar.y = 0.5, n.start = 100)

  z <- cbind(1, mls(dgp$y, 1:2, 1))
  colnames(z) <- c("Intercept", "y1", "y2")
  X <- mls(dgp$x, 0:23, 12)

  lstr_mod_plain <- midas_lstr_plain(dgp$y, X, z, nnbeta,
    start_lstr = c(1.5, 1, 1, 1), start_x = c(2, 4), start_z = c(1, 0.5, 0),
    method = "Nelder-Mead", control = list(maxit = 5000)
  )
  lstr_mod_nlpr <- midas_nlpr(y ~ mlsd(x, 0:23, y, nnbeta) + mlsd(y, 1:2, y),
    data = dgp,
    start = list(
      x = list(
        lstr = c(1.5, 1, 1, 1),
        r = c(2, 4)
      ),
      y = c(0.5, 0),
      `(Intercept)` = 1
    )
  )
  list(dgp = dgp, plain = lstr_mod_plain, nlpr = lstr_mod_nlpr)
}

sim_lstr <- lapply(c(250, 500, 1000, 2000, 5000, 10000, 20000), do_n_lstr)

do_n_mmm <- function(n) {
  dgp <- midas_mmm_sim(n, m = 12, theta = nnbeta(c(2, 4), 24), intercept = 1, pmmm = c(1.5, 1), ar.x = 0.9, ar.y = 0.5, n.start = 100)

  z <- cbind(1, mls(dgp$y, 1:2, 1))
  colnames(z) <- c("Intercept", "y1", "y2")
  X <- mls(dgp$x, 0:23, 12)

  mmm_plain <- midas_mmm_plain(dgp$y, X, z, nnbeta,
    start_mmm = c(1.5, 1), start_x = c(2, 4), start_z = c(1, 0.5, 0),
    method = "Nelder-Mead", control = list(maxit = 5000)
  )

  mmm_nlpr <- midas_nlpr(y ~ mlsd(x, 0:23, y, nnbeta) + mlsd(y, 1:2, y),
    data = dgp,
    start = list(
      x = list(
        mmm = c(1.5, 1),
        r = c(2, 4)
      ),
      y = c(0.5, 0),
      `(Intercept)` = 1
    ),
    method = "Nelder-Mead", control = list(maxit = 5000)
  )

  list(dgp = dgp, plain = mmm_plain, nlpr = mmm_nlpr)
}

sim_mmm <- lapply(c(250, 500, 1000, 2000, 5000, 10000, 20000), do_n_mmm)