This is a simple program to calculate the R_0 of COVID-19 for each of the 5 districts in Penang, Malaysia. Namely, Timur Laut, Barat Daya, Seberang Perai Utara, Seberang Perai Tengah and Seberang Perai Selatan. The code is written in a way that makes it easy to use for any other state in Malaysia. If you have python3 with some basic packages installed (e.g. NumPy), it is easy to modify the program for any state in Malaysia (see Usage below). In due time, we will attach an overleaf link for a detailed decription of how R_0 is calculated.
Figure 1
Figure 1 shows the evolution over time of R_0 vs daily cases per 100k population for each district (TL, BD, SPU, SPT, SPS) in Penang, Malaysia. Notice towards the end of May/early June how the dots shift from green to yellow/red. This coincides with the spike in COVID-19 across Malaysia during the holiday season. Figure 1 was produced from R_0 values computed using r0_model.py.
I am grateful to Eric Cooper for many insightful conversations on COVID-19. I would also like to thank Kartik Chandra for his help with Git. Thanks to Daniel Comber-Todd for his suggestion of using R_0 vs daily new cases per 100k population as a metric for assessing the severity of COVID-19.
The facebook page Penang Lawan COVID-19 publishes daily reports of COVID-19 cases in Penang by district. I entered data from loc. cit. by hand into an excel file and used this as my dataset.
To see a list of all optional command line arguments, as well as a description of what these arguments mean, do:
python3 r0_model.py --h
python3 r0_model.py input_file output_file
The file params.py contains several parameters that you may modify for your own needs. If you want to use this for another state in Malaysia (e.g. Selangor or Kedah), change REGION to your state and SUBREGIONS to the list of districts (or daerah in Malay) for that state. Make sure that the strings in SUBREGIONS are spelled (lower/upper case, spacing between words) correctly. In other words, they agree with what you have in the head of your .csv file.
Generally speaking, it is incredibly difficult to obtain COVID-19 data in Malaysia. The government publishes daily reports of COVID-19 statistics such as new cases, deaths and test positivity rates. However, to my knowledge there is no public repository of data stored in a systematic way in .xlsx or .csv files that are conducive to modelling. I would be more than happy if volunteers could help enter this data, in addition to other things such as daily deaths, test positivity rates, etc. If you wish to help, please email me.