Changed naming of matrix A to use "aggregation matrix" everywhere.

NEWS.md

@@ -1,6 +1,6 @@
 # bayesRecon 0.3.1
 
-* IMPORTANT CHANGE IN THE API OF THE `reconc_*` functions: they now require the aggregating matrix A and not the summing matrix S.
+* IMPORTANT CHANGE IN THE API OF THE `reconc_*` functions: they now require the aggregation matrix A and not the summing matrix S.
 
 * The examples section of the `reconc_TDcond` now contains an example showing how to handle the case of an unbalanced hierarchy. 
 

R/hierarchy.R

@@ -413,7 +413,7 @@ get_reconc_matrices <- function(agg_levels, h) {
 }
 
 
-# Get the aggregating matrix Au of the sub-hierarchy composed just by the uppers 
+# Get the aggregation matrix Au of the sub-hierarchy composed just by the uppers 
 .get_Au <- function(A, lowest_rows=NULL) {
 
   if (is.null(lowest_rows)) lowest_rows = .lowest_lev(A)

R/reconc_MCMC.R

@@ -208,7 +208,7 @@ reconc_MCMC <- function(A,
 # @title Compute acceptance probability
 # @param b proposal state
 # @param b0 current state
-# @param A aggregating matrix
+# @param A aggregation matrix
 # @param distr list of strings specifying the distribution of each variable
 # @param params list of the parameters of the distributions
 # @return the acceptance probability alpha

R/reconc_TDcond.R

@@ -219,7 +219,7 @@
 #'                  5, 0,  0, var_b5), nrow=4, byrow = TRUE)
 #' fc_upper<- list(mu=mu, Sigma=Sigma)
 #' 
-#' # We also need to update the aggregating matrix
+#' # We also need to update the aggregation matrix
 #' A <- matrix(c(1,1,1,1,1,
 #'               1,1,0,0,0,
 #'               0,0,1,1,0,

vignettes/bayesRecon.Rmd

@@ -154,15 +154,15 @@ for (l in seq_along(train.agg)) {
 ```
 
 
-Reconciliation requires the aggregating matrix $\mathbf{A}$, which we obtain using the function `get_reconc_matrices`. 
+Reconciliation requires the aggregation matrix $\mathbf{A}$, which we obtain using the function `get_reconc_matrices`. 
 It requires: 
 
 *   the aggregation factors of the hierarchy, which in this example are $\{2, 3, 4, 6, 12\}$;
 *   the length of the forecasting horizon at the  bottom level, which is 12 in this example.
 
-``` {r aggregatingMatrix}
+``` {r aggregationMatrix}
 recon.matrices <- bayesRecon::get_reconc_matrices(agg_levels = c(2, 3, 4, 6, 12), h = 12)
-# Aggregating matrix
+# Aggregation matrix
 A <- recon.matrices$A 
 ```
 
@@ -279,7 +279,7 @@ for (level in train.agg) {
 
 Using the function `get_reconc_matrices`,  we get  matrix $\mathbf{A}$.
 
-```{r m3-rmat, dpi=300, out.width = '70%', fig.align='center', fig.cap="**Figure 4**: M3 - The aggregating matrix A (red=1, yellow=0).", fig.dim = c(8, 8)}
+```{r m3-rmat, dpi=300, out.width = '70%', fig.align='center', fig.cap="**Figure 4**: M3 - The aggregation matrix A (red=1, yellow=0).", fig.dim = c(8, 8)}
 rmat <- get_reconc_matrices(agg_levels = c(2, 3, 4, 6, 12), h = 18)
 
 par(mai = c(1,1,0.5,0.5))
@@ -392,7 +392,7 @@ for (s in infantMortality) {
 
 Now we build the $\mathbf{A}$ matrix.
 
-```{r infants-s, dpi=300, out.width = '70%', fig.align='center', fig.cap="**Figure 6**: Infants mortality - The aggregating matrix A (red=1, yellow=0).", fig.dim = c(8, 8)}
+```{r infants-s, dpi=300, out.width = '70%', fig.align='center', fig.cap="**Figure 6**: Infants mortality - The aggregation matrix A (red=1, yellow=0).", fig.dim = c(8, 8)}
 # we have 16 bottom time series, and 11 upper time series
 A <- matrix(data = c(1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
                      1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
@@ -444,4 +444,4 @@ upper_mu_reconc
 ```
 
 # References
-<div id="refs"></div>
+<div id="refs"></div>

vignettes/reconciliation_properties.Rmd

@@ -60,7 +60,7 @@ n_b <- 5
 n_u <- 1
 n <- n_b + n_u   
 
-A <- matrix(1, ncol = n_b, nrow = n_u)  # aggregating matrix
+A <- matrix(1, ncol = n_b, nrow = n_u)  # aggregation matrix
 
 # Actual values:
 actuals <- data.frame(extr_mkt_events)  # convert to data frame