 Thank you. Thank you very much, Anna. So we'll take five minutes of question because then we should start on time the next session and there's a break in between, but I mean, one interesting question in the Q&A is basically a twist around your amplification effect here. So would it be possible with your data to check whether doves become more dovish and hoax become more hawkish in periods of high uncertainty? So that's a very interesting question. In principle, yes, because and we do have some results at the individual level. So the data is granular. The data allows we observe statements at each individual level by section of the meeting and we have constructed individual level policy preference measures as well as uncertainty measures as well as sentiment measures. So in principle, yes, we can see whether extreme uncertainty would lead to more divergence of views and separation essentially into stronger camps. We could do that. Thank you. Another question is, you know, why do you not use the voting record to obtain information about policy makers preferences in transcripts? Because there's a lot of, you know, discussion which may give a wrong indication about the true preferences. So somehow the vote is really the final call. So, so actually, I'm not quite sure about vote being the more interesting measure. We know that voting is that the sense have been incredibly rare over the sample that we study that so, so there will not be a lot of, you know, heterogeneity there. And what we show actually what is very interesting about our preference measure is that it is not only predictive of, you know, current market based surprises. It is not only predictive of Romer and Romer shops. It is actually predicting monetary policy stands going forward several quarters ahead. This is suggesting that the discussions in the transcripts, you know, are not just some misleading conversations, they actually present a lot of information about the intended path of the monetary policy going forward and are really forward looking, which, which we found very interesting. In the context of us trying to understand where monetary policy shocks actually come from. Thank you. And if I may, you know, abuse my chair privilege here. I mean, why do you not use the part of the transcripts related to the financial markets presentation at the beginning, because one could assume that it's exactly, you know, one financial market is volatile and you have correlations breaking up liquidity being disrupted that this can really have an impact actually on the Fed reaction function and you know that financial market practitioners have this theory of the Fed puts yeah. So, why do you leave that out. Yes, so, so we could extend it to include the markets section and in fact historically. We did have we were constructing our indices using that information to the only they wouldn't change our conclusions. The only reason why we are not doing it right now is that the structure of the market section has changed over time. And so we think that the important elements of the market deliberations about what happened in financial markets will find their way to the discussion of economic conditions as well. So we find strong correlation between you know what would happen in the markets section with what actually finds its way to the economy discussion. Okay, so it's a kind of warm up for the economy discussion. Yeah. Last question maybe in order to leave time for the break is you mentioned that it is hard or almost impossible to distinguish pure uncertainty from risk. What about risk aversion. Yeah. So, and the question says it could be that a given amount of uncertainty or risk can lead to longer discussions in the FMC meetings when policymakers risk aversion is higher than usual risk aversion. No, absolutely. So we are not this, we are not really taking a stance or of the risk aversion of policymakers, but with our sentiment measures, or at least in terms of inflation what we establish is that a lot of the expression of uncertainty that happens in the meeting is actually the expression of a worry about a really undesirable outcomes. And you could think about this as a proxy for risk aversion of policymakers. And then obviously our subsequent measures of policy preferences are showing that that this type of uncertainty that is related with undesirable outcomes or concerns has strong effects on actual policy preferences. Thank you. Thank you very much. So on that note, I will close the session and thank you very much for the paper and the great, the great Q&A.