bootstrapping bloomberg

Does anyone know the zero rate here at -0.23022 is derived? I have tried (1+0.0056*0.503)*(1+-0.00232*0.086)=(1+?^(1/0.589). Solving for ? gives me -0.002344. I have tried simple and compounded interest. I cannot get this to match. can anyone provide clues?

This is not exactly an answer to your question, but I have found that for practical purpose it is best to use directly the discount factors (last column on the screen), which you can export to Excel and interpolate according to your prefered method for specific maturities.

Beware that the curve reference date (the date for which the discount factor is 1) is the Settle Date (3rd field on the upper left of the screen).

• I don't wish to do that. I would like to derive zero rates from par and then use those to derive df. Also I am not trying to do any interpolation for specific maturities. I am just trying to get this for standard tenors like what's shown on that screen.
– Akk
Jan 22 '19 at 0:04
• Not sure I understand. You can compute the zero rates from the discount factors. Jan 22 '19 at 7:33
• ok how do you derive df from par rates? I tried doing it that way too and I couldnt get it. I know I can always download df directly by just copying them from the screen. i am trying to use the par rates to derive either one and then get the other.
– Akk
Jan 22 '19 at 9:00
• Ok. I don’t think they use a consistent convention across maturities (for instance the first tenor clearly uses simple interest but not the longer tenors). But DF are prices so there is no convention here and they can be used as is. Also the interpolation scheme they use is “Step Forward (Cont)” meaning that the instantaneous forward is a step function which is equivalent to saying that log DF are linearly interpolated. So their bootstrapping probably works directly on DF, and the “zero rate” are only computed from the DF after these have been bootstrapped. Jan 22 '19 at 10:29
• No that is not correct. DF/ zero rates come from par rates. They're not pulled out from thin air. How do they come about? The bootstrapping is done on par rates. basically what I am trying to ask is very similar to what was done on quant.stackexchange.com/questions/36441/… but that method didn't work in this case. Is anyone able to take a stab at this please.
– Akk
Jan 23 '19 at 8:17

can you add to your screenshot a bit further to the left? i am guessing that 2nd row is a 6m fra, in which case to work out the discount factor, you probably need an interpolation assumption as to how the 1m is gotten