I read this news (link below) yesterday that indicates that BRL might be very expensive according to three models from Deutsche Bank. Though I don't have access to those models metrics, it got my attention because that information doesn't really fit very well with some simple models I run. I know there are lots of ways to model FX and as everyone might know its not a very easy market to figure out, specially for EM, where lots of feedback loops between country risk and FX rate happens and you can have long and short term models. Another point, specially for the PPP metric is the timeframe chosed. Anyway, just decided to post my findings here because they contrast with the DB picture. They might be right but here is my view: Business Insider (Australia) link for the news: https://www.businessinsider.com.au/currency-valuations-september-deutsche-bank-2017-10 That's DB chart... My Behavioral model based on Commodities and CDS (YoY %) PPP metrics based on C