I'm currently trying to get the implied volatility of a vanilla Euro floor with maturity 1Y with data from bloom.
I have the price ( which is not supposed to take into account the first floorlet which has already fixed, but takes into account only the three remaining optional floorlet.) of the floor and all the convetions which are supposed to be correct ( Act/360, floor starting 2 open days after fixing, etc...). I manage to get this implied volatility for somes strikes and even for all strikes some day, but quite often I have floors with strike ( say 4%) that I cannot get the implied vol from.
It turns out actually that my forward rates are too low and so I get a floor too high for the quoted price. Even a zero volatility get me a price above the market quote. I also price the same floorlet in Bloom and get a price above the market quote even for zero vol. So the problem I think lies in my EUR3M forward curve.The worst thing is that I was not even using EONIA curve for discounting my floorlet coupon so actually I should be even further from the quoted price.
I construct my forward curve, with cash ( euribor 3M), futures with 3M maturity ( that I turn into FRA) up to maturity 2 years and then swaps. This is the way Bloom is doing. A priori OIS should not impact in this case my curve for maturity below 2 years. So I'm wondering wha'ts wrong with my forward curve? Should I remove the futures and keep only swaps? Is it the interpolation of rates? Should I smoothe my curve? Should I use in some ways EONIA below 2y for boostrapping? Thank you for your help