Paper on Forecasting Ambulance Call-outs by Bahman Rostami-Tabar and Rob J Hyndman
data
contains raw data and cleaned/wrangled data (stored as rds) ingts
andtsibble
formats.rscripts
contains all R scripts used in experiments and in final versionmanuscript
contains our main paperimg
contains figures for the paper not created using R
Run the following scripts in this order
- Storage folder: Edit following line to wherever your Google drive folder is
storage_folder <- "/home/hyndman/GDrive/Incidents_project/"
- Prepare data:
rscript/data/incidents.R
- Generate forecasts and evaluation summaries:
rscript/hts/forecast_hts.R
- Hierarchical dynamic models for multivariate times series of counts
- paperpile
- Dynamic Models for Time Series of Counts with a Marketing Application
- https://paperpile.com/shared/IB0z2K
- https://cran.r-project.org/web/packages/glarma/vignettes/glarma.pdf
- https://www.intlpress.com/site/pub/pages/journals/items/sii/content/vols/0007/0004/a011/
- https://paperpile.com/shared/P1EJ9e
- https://cran.r-project.org/web/packages/tscount/vignettes/tsglm.pdf
- https://cran.r-project.org/web/packages/glarma/vignettes/glarma.pdf
- https://cran.r-project.org/web/packages/GAS/index.html
- weather data: http://data.ceda.ac.uk/badc/ukmo-midas-open/data/uk-daily-temperature-obs/dataset-version-202007/south-glamorgan/01272_cardiff-bute-park/qc-version-1
- https://cran.r-project.org/web/views/TimeSeries.html
- check: https://bbolker.github.io/mixedmodels-misc/glmmFAQ.html
- tscount: https://cran.r-project.org/web/packages/tscount/vignettes/tsglm.pdf
- https://pipiras.sites.oasis.unc.edu/timeseries/Nonlinear_2_-_Count_time_series_-_Menu.html
- https://hidda.github.io/forecasting/articles/extra/CHILI_tscount.html
- https://www.youtube.com/watch?v=6mIUmAUj0I0
- https://masalmon.eu/2017/02/12/wikideaths1_ts/
- file:///Users/bahmanrostami-tabar/Downloads/2015_31_report.pdf
- https://hidda.github.io/forecasting/
- https://hidda.github.io/forecasting/articles/CHILI_glarma.html