re-resubmit v1.0.0 to cran

ygeunkim · Nov 8, 2023 · a087f85 · a087f85
1 parent 952ee6a
commit a087f85
Show file tree

Hide file tree

Showing 28 changed files with 120 additions and 122 deletions.
diff --git a/CRAN-SUBMISSION b/CRAN-SUBMISSION
@@ -1,3 +1,3 @@
 Version: 1.0.0
-Date: 2023-11-07 11:22:14 UTC
-SHA: fa382376cbbee3dfdf567b1dcdbeb87db0dfebe9
+Date: 2023-11-08 05:42:19 UTC
+SHA: 952ee6ac28389258abdcd3908b7b821b6f5587da
diff --git a/NAMESPACE b/NAMESPACE
@@ -152,7 +152,6 @@ export(choose_bvar)
 export(choose_bvhar)
 export(choose_ssvs)
 export(choose_var)
-export(compute_cov)
 export(compute_dic)
 export(compute_logml)
 export(conf_fdr)

diff --git a/R/RcppExports.R b/R/RcppExports.R
@@ -300,7 +300,7 @@ estimate_var <- function(x, y, method) {
 #' \deqn{\hat{\Sigma}_e = \frac{1}{s - k} (Y_0 - \hat{A} X_0)^T (Y_0 - \hat{A} X_0)}
 #' 
 #' @references Lütkepohl, H. (2007). *New Introduction to Multiple Time Series Analysis*. Springer Publishing.
-#' @export
+#' @noRd
 compute_cov <- function(z, num_design, dim_design) {
     .Call(`_bvhar_compute_cov`, z, num_design, dim_design)
 }
@@ -343,7 +343,7 @@ VARcoeftoVMA <- function(var_coef, var_lag, lag_max) {
 #' \deqn{W_1 = W_0 B_1 (W_1^T = B_1^T W_0^T)}
 #' \deqn{W_2 = W_1 B_1 + W_0 B_2 (W_2^T = B_1^T W_1^T + B_2^T W_0^T)}
 #' \deqn{W_j = \sum_{j = 1}^k W_{k - j} B_j (W_j^T = \sum_{j = 1}^k B_j^T W_{k - j}^T)}
-#' 
+#' @return VMA coefficient of k(lag-max + 1) x k dimension
 #' @references Lütkepohl, H. (2007). *New Introduction to Multiple Time Series Analysis*. Springer Publishing.
 #' @export
 VARtoVMA <- function(object, lag_max) {
@@ -463,7 +463,7 @@ VHARcoeftoVMA <- function(vhar_coef, HARtrans_mat, lag_max, month) {
 #' 
 #' Observe that
 #' \deqn{B = \tilde{T}^T \Phi}
-#' 
+#' @return VMA coefficient of k(lag-max + 1) x k dimension
 #' @references Lütkepohl, H. (2007). *New Introduction to Multiple Time Series Analysis*. Springer Publishing.
 #' @export
 VHARtoVMA <- function(object, lag_max) {
@@ -880,14 +880,14 @@ sim_mstudent <- function(num_sim, df, mu, sig, method) {
 #' @param mat_scale_u First scale matrix
 #' @param mat_scale_v Second scale matrix
 #' @details
-#' Consider s x m matrix \eqn{Y_1, \ldots, Y_n \sim MN(M, U, V)} where M is s x m, U is s x s, and V is m x m.
+#' Consider n x k matrix \eqn{Y_1, \ldots, Y_n \sim MN(M, U, V)} where M is n x k, U is n x n, and V is k x k.
 #' 
 #' 1. Lower triangular Cholesky decomposition: \eqn{U = P P^T} and \eqn{V = L L^T}
 #' 2. Standard normal generation: s x m matrix \eqn{Z_i = [z_{ij} \sim N(0, 1)]} in row-wise direction.
 #' 3. \eqn{Y_i = M + P Z_i L^T}
 #' 
 #' This function only generates one matrix, i.e. \eqn{Y_1}.
-#' 
+#' @return One n x k matrix following MN distribution.
 #' @export
 sim_matgaussian <- function(mat_mean, mat_scale_u, mat_scale_v) {
     .Call(`_bvhar_sim_matgaussian`, mat_mean, mat_scale_u, mat_scale_v)
@@ -916,13 +916,13 @@ sim_iw_tri <- function(mat_scale, shape) {
 #' @details
 #' Consider \eqn{\Sigma \sim IW(\Psi, \nu)}.
 #' 
-#' 1. Upper triangular Bartlett decomposition: m x m matrix \eqn{Q = [q_{ij}]} upper triangular with
+#' 1. Upper triangular Bartlett decomposition: k x k matrix \eqn{Q = [q_{ij}]} upper triangular with
 #'     1. \eqn{q_{ii}^2 \chi_{\nu - i + 1}^2}
 #'     2. \eqn{q_{ij} \sim N(0, 1)} with i < j (upper triangular)
 #' 2. Lower triangular Cholesky decomposition: \eqn{\Psi = L L^T}
 #' 3. \eqn{A = L (Q^{-1})^T}
 #' 4. \eqn{\Sigma = A A^T \sim IW(\Psi, \nu)}
-#' 
+#' @return One k x k matrix following IW distribution
 #' @export
 sim_iw <- function(mat_scale, shape) {
     .Call(`_bvhar_sim_iw`, mat_scale, shape)
@@ -941,10 +941,11 @@ sim_iw <- function(mat_scale, shape) {
 #' Consider \eqn{(Y_i, \Sigma_i) \sim MIW(M, U, \Psi, \nu)}.
 #' 
 #' 1. Generate upper triangular factor of \eqn{\Sigma_i = C_i C_i^T} in the upper triangular Bartlett decomposition.
-#' 2. Standard normal generation: s x m matrix \eqn{Z_i = [z_{ij} \sim N(0, 1)]} in row-wise direction.
+#' 2. Standard normal generation: n x k matrix \eqn{Z_i = [z_{ij} \sim N(0, 1)]} in row-wise direction.
 #' 3. Lower triangular Cholesky decomposition: \eqn{U = P P^T}
 #' 4. \eqn{A_i = M + P Z_i C_i^T}
-#' 
+#' @return List of MN and IW matrices.
+#' Multiple samples are column-stacked.
 #' @export
 sim_mniw <- function(num_sim, mat_mean, mat_scale_u, mat_scale, shape) {
     .Call(`_bvhar_sim_mniw`, num_sim, mat_mean, mat_scale_u, mat_scale, shape)

diff --git a/R/bvar-hierarchical.R b/R/bvar-hierarchical.R
@@ -19,6 +19,7 @@
 #'   delta = rep(1, 3),
 #'   eps = 1e-04 # eps = 1e-04
 #' )
+#' @return `bvharpriorspec` object
 #' @references Giannone, D., Lenza, M., & Primiceri, G. E. (2015). *Prior Selection for Vector Autoregressions*. Review of Economics and Statistics, 97(2).
 #' @order 1
 #' @export

diff --git a/R/criteria.R b/R/criteria.R
@@ -178,7 +178,7 @@ logLik.bvharmn <- function(object, ...) {
 #' @param ... not used
 #' @details 
 #' Let \eqn{\tilde{\Sigma}_e} be the MLE
-#' and let \eqn{\hat{\Sigma}_e} be the unbiased estimator (from [compute_cov()] and the member named `covmat`) for \eqn{\Sigma_e}.
+#' and let \eqn{\hat{\Sigma}_e} be the unbiased estimator (`covmat`) for \eqn{\Sigma_e}.
 #' Note that
 #' 
 #' \deqn{\tilde{\Sigma}_e = \frac{s - k}{s} \hat{\Sigma}_e}
@@ -271,17 +271,16 @@ FPE <- function(object, ...) {
 #' @param ... not used
 #' @details 
 #' Let \eqn{\tilde{\Sigma}_e} be the MLE
-#' and let \eqn{\hat{\Sigma}_e} be the unbiased estimator (from [compute_cov()] and the member named `covmat`) for \eqn{\Sigma_e}.
+#' and let \eqn{\hat{\Sigma}_e} be the unbiased estimator (`covmat`) for \eqn{\Sigma_e}.
 #' Note that
 #' 
 #' \deqn{\tilde{\Sigma}_e = \frac{s - k}{n} \hat{\Sigma}_e}
 #' 
 #' Then
 #' 
 #' \deqn{FPE(p) = (\frac{s + k}{s - k})^m \det \tilde{\Sigma}_e}
-#' 
+#' @return FPE value.
 #' @references Lütkepohl, H. (2007). *New Introduction to Multiple Time Series Analysis*. Springer Publishing.
-#' 
 #' @export
 FPE.varlse <- function(object, ...) {
   compute_fpe(object)
@@ -303,7 +302,7 @@ FPE.vharlse <- function(object, ...) {
 #' @param ... not used
 #' @details 
 #' Let \eqn{\tilde{\Sigma}_e} be the MLE
-#' and let \eqn{\hat{\Sigma}_e} be the unbiased estimator (from [compute_cov()] and the member named `covmat`) for \eqn{\Sigma_e}.
+#' and let \eqn{\hat{\Sigma}_e} be the unbiased estimator (`covmat`) for \eqn{\Sigma_e}.
 #' Note that
 #' 
 #' \deqn{\tilde{\Sigma}_e = \frac{s - k}{n} \hat{\Sigma}_e}
@@ -406,7 +405,7 @@ HQ.logLik <- function(object, ...) {
 #' @param ... not used
 #' @details 
 #' Let \eqn{\tilde{\Sigma}_e} be the MLE
-#' and let \eqn{\hat{\Sigma}_e} be the unbiased estimator (from [compute_cov()] and the member named `covmat`) for \eqn{\Sigma_e}.
+#' and let \eqn{\hat{\Sigma}_e} be the unbiased estimator (`covmat`) for \eqn{\Sigma_e}.
 #' Note that
 #' 
 #' \deqn{\tilde{\Sigma}_e = \frac{s - k}{n} \hat{\Sigma}_e}
@@ -470,9 +469,9 @@ HQ.bvharmn <- function(object, ...) {
     AIC()
 }
 
-#' Deviance Information Criterion
+#' Deviance Information Criterion of Multivariate Time Series Model
 #' 
-#' Generic function that computes DIC.
+#' Compute DIC of BVAR and BVHAR.
 #' 
 #' @param object Model fit
 #' @param ... not used
@@ -482,10 +481,7 @@ compute_dic <- function(object, ...) {
   UseMethod("compute_dic", object)
 }
 
-#' Deviance Information Criterion of Multivariate Time Series Model
-#' 
-#' Compute DIC of BVAR and BVHAR
-#' 
+#' @rdname compute_dic
 #' @param object Model fit
 #' @param n_iter Number to sample
 #' @param ... not used

diff --git a/R/stable-process.R b/R/stable-process.R
@@ -30,9 +30,8 @@ is.stable <- function(x, ...) {
 #' \deqn{\det(I_m - A z) = 0}
 #' 
 #' where \eqn{A} is VAR(1) coefficient matrix representation.
-#' 
+#' @return Numeric vector.
 #' @references Lütkepohl, H. (2007). *New Introduction to Multiple Time Series Analysis*. Springer Publishing.
-#' 
 #' @export
 stableroot.varlse <- function(x, ...) {
   compute_var_stablemat(x) %>% 
@@ -53,9 +52,8 @@ stableroot.varlse <- function(x, ...) {
 #' \deqn{\det(I_m - A z) \neq 0}
 #' 
 #' for \eqn{\lvert z \rvert \le 1}.
-#' 
+#' @return logical class
 #' @references Lütkepohl, H. (2007). *New Introduction to Multiple Time Series Analysis*. Springer Publishing.
-#' 
 #' @export
 is.stable.varlse <- function(x, ...) {
   all(stableroot(x) < 1)

diff --git a/R/tuning.R b/R/tuning.R
@@ -598,6 +598,7 @@ choose_bayes <- function(bayes_bound = bound_bvhar(),
 #' In case of \eqn{(\omega_{0ij}, \omega_{1ij})},
 #' \deqn{\omega_{kij} = c_k = \hat{VAR(OLS)}}
 #' similarly.
+#' @return `ssvsinput` object
 #' @references 
 #' George, E. I., & McCulloch, R. E. (1993). *Variable Selection via Gibbs Sampling*. Journal of the American Statistical Association, 88(423), 881–889.
 #' 

diff --git a/cran-comments.md b/cran-comments.md
@@ -1,3 +1,9 @@
+## Re-resubmission v1.0.0
+This is second resubmission. Comments are included in italics.
+
+- *We still see: Missing Rd-tags ... Since you are using 'roxygen', please add a @return tag in the corresponding .R-file and re-roxygenize() your .Rd-files.*\
+**Response** We have added `@return` to each mentioned document.
+
 ## Resubmission
 This is a resubmission. Comments are included in italics.
 
@@ -15,3 +21,10 @@ This is a resubmission. Comments are included in italics.
 0 errors | 0 warnings | 1 note
 
 * This is a new release.
+
+## revdepcheck results
+
+We checked 0 reverse dependencies, comparing R CMD check results across CRAN and dev versions of this package.
+
+ * We saw 0 new problems
+ * We failed to check 0 packages
diff --git a/man/AIC.varlse.Rd b/man/AIC.varlse.Rd
diff --git a/man/BIC.varlse.Rd b/man/BIC.varlse.Rd
diff --git a/man/FPE.varlse.Rd b/man/FPE.varlse.Rd
diff --git a/man/HQ.varlse.Rd b/man/HQ.varlse.Rd
diff --git a/man/VARtoVMA.Rd b/man/VARtoVMA.Rd
diff --git a/man/VHARtoVMA.Rd b/man/VHARtoVMA.Rd
diff --git a/man/choose_ssvs.Rd b/man/choose_ssvs.Rd
diff --git a/man/compute_cov.Rd b/man/compute_cov.Rd
diff --git a/man/compute_dic.Rd b/man/compute_dic.Rd