The methodology underlying the forecasts presented on this website
is described in detail in our
working paper. It is also available at
NBER Working Paper 27248 and
CEPR Working Paper 14790. A blog post is available
Here are the
that illustrate our method using the 04/18/2020 sample.
The forecasts are based on a dynamic panel data model. At the core
of our model is a specification that assumes that the growth rate of
active infections can be represented by autoregressive fluctuations
around a downward sloping deterministic trend function with a break.
used to generate the forecasts is obtained from the Center for
Systems Science and Engineering (CSSE) at Johns Hopkins University.
Results as of
(updating weekly on Sundays).
Click on a link in the grid below to see graphs for a specific
forecast origin and horizon.