Corona Deaths

Corona Deaths#

In this project we collect and/or compute death rates before and during the Corona pandemic in Germany. You should read Dates and Times before you start.

Get some Data#

We would like to have monthly death rates for an as long as possible period of time including very recent data.

Task: Download relevant data from Federal Statistical Office (Statistisches Bundesamt) in CSV format:

For each file write a short note on its content.

Solution:

# your notes

Task: Use your favorit spreadsheet tool to compile following CSV files from the downloaded files:

  • inhabitants-yearly.csv with columns year, FRG (inhabitants FRG), GDR (inhabitants GDR, 0 from 1990 on)

  • inhabitants-quarterly.csv with colums date, inhabitants

  • deaths-monthly.csv with columns year, months (numeric 1…12), men, women

Load Data#

We want to use dates as index for data frames. Numbers of inhabitants are related to precise timestamps (end of year or quarter). Numbers of deaths are related to periods (month).

Task: Read in the three CSV files. Use DatetimeIndex and PeriodIndex for data frames and series. In the end you should have two series:

  • inhabitants with index date (timestamp of last day in year or quarter),

  • deaths with index date (monthly period aligned at last day of month).

Solution:

# your solution

Death Rates#

For calculating monthly death rates we have to get the number of inhabitants on a monthly basis, i.e., the mean number of inhabitants per month. If we would have daily values for the number of inhabitants we could simply calculate the mean. But resolution is much coarser. Thus, we have to use (linear) interpolation. A good replacement for the monthly mean is the (interpolated) value at the 15th of the month.

Task: Use resampling to get interpolated number of inhabitants at the 15th of each month. From these values construct a series with period index (in analogy to the deaths series’ index). Hint: instead of (integer) index based linear interpolation you may want to use timestamp based interpolation (see docs).

# your solution

Task: Calculate monthly death rates and plot results with Series.plot().

# your solution
Contents