Fixes for oesg

Author: config-i1

R/oesg.R

@@ -803,7 +803,7 @@ oesg <- function(y, modelA="MNN", modelB="MNN", persistenceA=NULL, persistenceB=
             # This is needed for the degrees of freedom calculation
             nParamB <- length(BValuesB$B);
 
-            if(length(BValuesA$B)>0 || length(BValuesB$B)>0){
+            if(length(BValuesA$B)>0 && length(BValuesB$B)>0){
                 # Run the optimisation
                 res <- nloptr(c(BValuesA$B,BValuesB$B), CF, lb=c(BValuesA$lb,BValuesB$lb), ub=c(BValuesA$ub,BValuesB$ub),
                               opts=list(algorithm=algorithm, xtol_rel=xtol_rel, maxeval=maxeval, print_level=print_level),
@@ -829,6 +829,9 @@ oesg <- function(y, modelA="MNN", modelB="MNN", persistenceA=NULL, persistenceB=
 
                 B <- res$solution;
             }
+            else{
+                B <- c(BValuesA$B,BValuesB$B);
+            }
         }
 
         ##### Deal with the fitting and the forecasts #####

vignettes/oes.Rmd

@@ -86,6 +86,7 @@ In this vignette we will use ETS(M,N,N) model as a base for the different parts
 
 We will use an artificial data in order to see how the functions work:
 ```{r artificialData}
+set.seed(41)
 y <- ts(c(rpois(20,0.25),rpois(20,0.5),rpois(20,1),rpois(20,2),rpois(20,3),rpois(20,5)))
 ```
 
@@ -315,9 +316,9 @@ oETSGModel2
 plot(oETSGModel2)
 ```
 
-Finally, the more flexible way to construct iETS model would be to do it in two steps: either using `oesg()` or `oes()` and then using the `es()` with the provided model in `occurrence` variable. But a simpler option is available as well:
+Finally, the more flexible way to construct iETS model would be to do it in two steps: either using `oesg()` or `oes()` and then using the `adam()` with the provided model in `occurrence` variable. But a simpler option is available as well:
 ```{r iETSGExample3}
-es(y, "MMN", occurrence="g", oesmodel="MMN", h=10, holdout=TRUE, silent=FALSE)
+adam(y, "MMN", occurrence="g", oesmodel="MMN", h=10, holdout=TRUE, silent=FALSE)
 ```