Merge pull request #10 from IDSIA/vignette_IJF

Vignette ijf

.Rbuildignore

@@ -9,3 +9,4 @@
 ^codecov\.yml$
 ^doc$
 ^Meta$
+^data-raw$

R/bayesRecon-package.R

@@ -21,7 +21,7 @@
 #'
 #' Corani, G., Azzimonti, D., Rubattu, N. (2023). *Probabilistic reconciliation of count time series*. \doi{10.1016/j.ijforecast.2023.04.003}.
 #'
-#' Zambon, L., Azzimonti, D. & Corani, G. (2022). *Efficient probabilistic reconciliation of forecasts for real-valued and count time series*. \doi{10.48550/arXiv.2210.02286}.
+#' Zambon, L., Azzimonti, D. & Corani, G. (2024). *Efficient probabilistic reconciliation of forecasts for real-valued and count time series*. \doi{10.1007/s11222-023-10343-y}.
 #'
 #' Zambon, L., Agosto, A., Giudici, P., Corani, G. (2023). *Properties of the reconciled distributions for Gaussian and count forecasts*. \doi{10.48550/arXiv.2303.15135}.
 #'

R/extr_mkt_events.R

@@ -0,0 +1,26 @@
+#' Extreme market events dataset
+#'
+#' Count time series of extreme market events in five economic sectors.
+#' The data refer to the trading days between 2004/12/31 and 2018/12/19 (3508 trading days in total).
+#' 
+#' The counts are computed by considering 29 companies included in the Euro Stoxx 
+#' 50 index and observing if the value of the CDS spread on a given day exceeds 
+#' the 90-th percentile of its distribution in the last trading year.
+#' The companies are divided in the following  sectors: Financial (FIN), Information 
+#' and Communication Technology (ICT), Manufacturing (MFG), Energy (ENG), and Trade (TRD). 
+#' 
+#' There are 6 time series: 
+#' - 5 bottom time series, corresponding to the daily counts for each sector
+#' - 1 upper time series, which is the sum of all the bottom (ALL)
+#'
+#' @format
+#' A multivariate time series of class \link[stats]{ts}.
+#' 
+#' @source 
+#' Zambon, L., Agosto, A., Giudici, P., Corani, G. (2023). *Properties of the reconciled distributions for Gaussian and count forecasts*. \doi{10.48550/arXiv.2303.15135}.
+#' 
+#' @references 
+#' Zambon, L., Agosto, A., Giudici, P., Corani, G. (2023). *Properties of the reconciled distributions for Gaussian and count forecasts*. \doi{10.48550/arXiv.2303.15135}.
+#'
+#' Agosto, A. (2022). *Multivariate Score-Driven Models for Count Time Series to Assess Financial Contagion*. \doi{10.2139/ssrn.4119895}
+"extr_mkt_events"

R/extr_mkt_events_basefc.R

@@ -0,0 +1,27 @@
+#' Base forecasts for the extreme market events dataset 
+#'
+#' Base forecasts for the `extr_mkt_events` dataset, computed using the model by 
+#' Agosto, A. (2022). *Multivariate Score-Driven Models for Count Time Series to Assess Financial Contagion*. \doi{10.2139/ssrn.4119895}. 
+#' 
+#' The predictive distribution for the bottom time series is a multivariate negative 
+#' binomial with a static vector of dispersion parameters and a time-varying vector 
+#' of location parameters following a score-driven dynamics. 
+#' The base forecasts for the upper time series are computed using a univariate version of this model.
+#' They are in-sample forecasts: for each training instant, they are computed for 
+#' time t+1 by conditioning on the counts observed up to time t. 
+#'
+#' @format
+#' A list `extr_mkt_events_basefc` containing
+#' \describe{
+#'    \item{`extr_mkt_events_basefc$mu`}{data frame of the base forecast means, for each day}
+#'    \item{`extr_mkt_events_basefc$size`}{data frame of the static base forecast size parameters}
+#' }
+#' 
+#' @source 
+#' Agosto, A. (2022). *Multivariate Score-Driven Models for Count Time Series to Assess Financial Contagion*. \doi{10.2139/ssrn.4119895}
+#' 
+#' @references 
+#' Agosto, A. (2022). *Multivariate Score-Driven Models for Count Time Series to Assess Financial Contagion*. \doi{10.2139/ssrn.4119895}
+#' 
+#' Zambon, L., Agosto, A., Giudici, P., Corani, G. (2023). *Properties of the reconciled distributions for Gaussian and count forecasts*. \doi{10.48550/arXiv.2303.15135}.
+"extr_mkt_events_basefc"

R/reconc.R

@@ -240,7 +240,7 @@
 #'print(rowMeans(samples_buis))
 #'
 #' @references
-#' Zambon, L., Azzimonti, D. & Corani, G. (2022). *Efficient probabilistic reconciliation of forecasts for real-valued and count time series*. \doi{10.48550/arXiv.2210.02286}.
+#' Zambon, L., Azzimonti, D. & Corani, G. (2024). *Efficient probabilistic reconciliation of forecasts for real-valued and count time series*. \doi{10.1007/s11222-023-10343-y}.
 #'
 #'
 #' @seealso

README.Rmd

@@ -236,8 +236,8 @@ Corani, G., Azzimonti, D., Augusto, J.P.S.C., Zaffalon, M. (2021). *Probabilisti
 Corani, G., Azzimonti, D., Rubattu, N. (2023). *Probabilistic reconciliation of count
 time series*. [DOI](https://doi.org/10.1016/j.ijforecast.2023.04.003)
 
-Zambon, L., Azzimonti, D. & Corani, G. (2022). *Efficient probabilistic reconciliation of forecasts
-for real-valued and count time series*. [DOI](https://doi.org/10.48550/arXiv.2210.02286)
+Zambon, L., Azzimonti, D. & Corani, G. (2024). *Efficient probabilistic reconciliation of forecasts
+for real-valued and count time series*. [DOI](https://doi.org/10.1007/s11222-023-10343-y)
 
 Zambon, L., Agosto, A., Giudici, P., Corani, G. (2023). *Properties of the reconciled distributions for Gaussian and count forecasts*. [DOI](https://doi.org/10.48550/arXiv.2303.15135)
 

README.md

@@ -256,9 +256,9 @@ Corani, G., Azzimonti, D., Rubattu, N. (2023). *Probabilistic
 reconciliation of count time series*.
 [DOI](https://doi.org/10.1016/j.ijforecast.2023.04.003)
 
-Zambon, L., Azzimonti, D. & Corani, G. (2022). *Efficient probabilistic
+Zambon, L., Azzimonti, D. & Corani, G. (2024). *Efficient probabilistic
 reconciliation of forecasts for real-valued and count time series*.
-[DOI](https://doi.org/10.48550/arXiv.2210.02286)
+[DOI](https://doi.org/10.1007/s11222-023-10343-y)
 
 Zambon, L., Agosto, A., Giudici, P., Corani, G. (2023). *Properties of
 the reconciled distributions for Gaussian and count forecasts*.

data-raw/extr_mkt_events.R

@@ -0,0 +1,9 @@
+library(readxl)
+
+dir_path <- dirname(rstudioapi::getSourceEditorContext()$path)
+setwd(dir_path)
+
+data        <- read_excel("NB_score_driven_forecasts.xlsx", sheet = "Data")
+extr_mkt_events <- ts(data)
+
+usethis::use_data(extr_mkt_events, overwrite = TRUE)

data-raw/extr_mkt_events_basefc.R

@@ -0,0 +1,12 @@
+library(readxl)
+
+dir_path <- dirname(rstudioapi::getSourceEditorContext()$path)
+setwd(dir_path)
+
+predictions <- read_excel("NB_score_driven_forecasts.xlsx", sheet = "Predictions")
+alpha       <- read_excel("NB_score_driven_forecasts.xlsx", sheet = "alpha", n_max = 1)
+
+extr_mkt_events_basefc <- list(mu   = data.frame(predictions),
+                               size = data.frame(1 / alpha))
+
+usethis::use_data(extr_mkt_events_basefc, overwrite = TRUE)