COVID-19

• real-time estimation of the Rt index
• analysis of COVID-19 data

Results

Results of the Rt computation are updated daily at 6PM, after Protezione Civile makes them available

The following data are updated:

• the full jupyther notebook notebooks/Rt_on_italian_national_data.ipynb
• the csv file with nuovi_positivi and $Rt$ values data/comupted/TD-covid19-ita-andamento-nazionale_rt.csv
• the excel file with nuovi_positivi and $Rt$ values data/comupted/TD-covid19-ita-andamento-nazionale_rt.xlsx
• the chart with the $Rt$ trend in images/TD_Rt_computation_MCMC_300dpi.png

Please note: file names are kept fixed (no dates in the names) so that they can be easily imported daily by third party applications. Refer to the commit date to check if the file has been updated.

Columns in the tables:

• data: the date, ISO format
• nuovi_positivi: raw value provided by Protezione Civile
• nuovi_positivi_Tikhonov: nuovi_positivi cleaned from noise, by Tikhonof regularization
• Rt: the Rt value computed with Markiv chain Monte Carlo, on the nuovi_positivi_Tikhonov
• Rt_interv_cred_min: min absolute value of cerdibility interval for the Rt value
• Rt_interv_cred_max: max absolute value of cerdibility interval for the Rt value

Third party Datasets

The sources and owners of the Datasets are cited in the notebooks. The static source files are in data/sources.

ISS weekly updates of the Rt value

Rt values weekly published by ISS are updated manually in the following file: data/sources/Rt_from_ISS.csv. The important dates in this file are in the column computation_time_range_descr which reports the period apopted by ISS for the computation of the Rt index.

That file is then processed to compute the reference data, which is set in the middle of the computation period. The processed file is available at: data/computed/Rt_from_ISS_processed.xlsx (or .csv) The processed file can be used to produce charts of the ISS Rt value.

Run with docker

• install docker
• install docker-compose
• run docker-compose up -d
• open browser and navigate to http://127.0.0.1:8888
• look at the the yaml