jensgrud

Definitely a report we did not want to hear. Just the the record though, the author is factually wrong about Denmark being one of the jurisdictions that "chose not to stamp out this virus and blithely tolerated high infection rates" and is probably referring to Sweden - see https://www.economist.com/the-world-ahead/2020/11/17/the-nor....

We actually wiped out an entire industry of mink farming in the name of stopping the spread of mutations, but that is a topic for an other thread (https://www.nytimes.com/2020/11/04/health/covid-mink-mutatio...).

I wrote my thesis last year comparing different RNNs against each other using this exact paper as baseline and basically concluded that you would be better off predicting the price yesterday than using their results. Authors did not respond when prompted for implementation details or comments.

Overall, concluded that amongst RNNs the GRU architecture proved most favorable but still would not outperform simple stochastic models of the financial industry toolbox.

You can check it out here https://github.com/jensgrud/financial-forecasting-lstm/tree/...