 Today I presented a paper which tries to use very cutting-edge machine learning tools in order to predict financial crisis. So monetary policy is very concerned as well with financial stability. Any central bank is very concerned with financial stability. And the problem so far is that we have not been very good at being able to have early warning indicators of crisis. And that's because there are very many different crises. They play out differently in different countries. There are many interactions between potential effects, between very many different variables. So we need powerful tools in order to have better early warning indicators. So what I do in my paper is that I use model aggregation techniques from machine learning in order to get the best possible predictor of crisis out of sample. And it works pretty well. So I found a lot of papers and presentations very interesting and stimulating in the conference so far. But I think maybe what strikes me most is that there were lots of presentations about the interaction between monetary policy and financial markets, about asset prices, financial stability. And I think that just shows that we need to do a lot more work in macro finance. And there are lots of very interesting research papers to be written on the interaction between central banks, monetary policy and also financial intermediaries and financial markets.