Re: [Quantlib-users] FuturesConvAdjustmentQuote issue in Quantlib

SourceForge Headquarters 1320 Columbia Street Suite 310 San Diego, CA 92101 +1 (858) 422-6466

Hi Tom,

Thank you! Now my calculations looks good as I see.

Regards,
Evgenia

-----Original Message-----
From: Tom Anderson [mailto:tw...@ur...] 
Sent: Monday, July 23, 2018 7:49 PM
To: Васильева Евгения Алексеевна
Cc: Arkadiy Naumov; qua...@li...
Subject: RE: [Quantlib-users] FuturesConvAdjustmentQuote issue in Quantlib

Hi Evgenia,

Those time calculations look correct to me. My code is basically the same:

       Date spot = iborIndex->valueDate(today);
       Date start = iborIndex->valueDate(expiryDate);
       Date end = iborIndex->fixingCalendar().advance(start, Period(3, TimeUnit::Months));
       DayCounter dayCounter = Actual360();
       double yf1 = dayCounter.yearFraction(spot, start);
       double yf2 = dayCounter.yearFraction(spot, end);
       double convexity = HullWhite::convexityBias(price, yf1, yf2, sigma, alpha);

The differences are trivial. There's no good reason for me to use an explicit period rather than the maturityDate method that you use, i don't know why i did!

tom

On Mon, 23 Jul 2018, Васильева Евгения Алексеевна wrote:

> Hi Arkadiy, thank you.
> 
> But I still cannot understand some issues:
>
> 1. the value date of the future
> It is like t. How do we calculate date to time? I try to get it from 
> Actual360: Time t = futDayCounter.yearFraction(settlementDate, imm); Is it right way?
>
> 2. the maturity date of the future's notional deposit
> It is like T. I try this function     Time T = futDayCounter.yearFraction(settlementDate, swFloatingLegIndex->maturityDate(imm));
> How do we calculate date to time in this case?
>
> Can you please help me.
>
> Thank you,
> Evgenia
>
>
>
> From: Arkadiy Naumov [mailto:ark...@gm...]
> Sent: Saturday, July 21, 2018 7:32 PM
> To: Васильева Евгения Алексеевна
> Cc: Tom Anderson; qua...@li...
> Subject: Re: [Quantlib-users] FuturesConvAdjustmentQuote issue in 
> Quantlib
>
> Evgeniya, I may be mistaken (others, please feel free to chime in), 
> but as far as I understand,  convexityBias function returns an 
> adjustment, not the underlying adjusted fwd rate. So you pass the 
> outcome of  convexityBias  into futures rate helper along with the 
> original quote, not instead of original quote
>
> On Thu, Jul 19, 2018 at 4:45 AM, Васильева Евгения Алексеевна <e.v...@1p...<mailto:e.v...@1p...>> wrote:
> Hi Tom,
>
> Could you please help in case as per below:
>
> Real futureQuote = 94.0;
>   Real a = 0.03;
>   Real sigma = 0.015;
>   Time t = 5.0;
>   Time T = 5.25;
> HullWhite::convexityBias(futureQuote,t,T,sigma,a); i got this example 
> from internet source. So there are being used Time, not Date. I can't have Dates as input for HullWhite::convexityBias You say inputs should be:
> 1. the value date of the future
> It is like t. How do we calculate date to time? I try to get it from 
> Actual360: Time t = futDayCounter.yearFraction(settlementDate, imm); Is it right way?
>
> 2. the maturity date of the future's notional deposit
> It is like T. I try this function     Time T = futDayCounter.yearFraction(settlementDate, swFloatingLegIndex->maturityDate(imm));
> Then i use HullWhite::convexityBias(97.54, t, T, 0.0754, 0.03) But 
> output is not even close to implied rate.
>
> Thank you,
> Evgenia
>
> -----Original Message-----
> From: Tom Anderson 
> [mailto:tw...@ur...<mailto:tw...@ur...>]
> Sent: Tuesday, July 17, 2018 7:07 PM
> To: Васильева Евгения Алексеевна
> Cc: 
> qua...@li...<mailto:qua...@li...
> eforge.net>
> Subject: Re: [Quantlib-users] FuturesConvAdjustmentQuote issue in 
> Quantlib Hi Evgenia,
>
> FuturesConvAdjustmentQuote is a wrapper around the HullWhite::convexityBias function:
>
> https://www.quantlib.org/reference/class_quant_lib_1_1_hull_white.html
> #af554e020436a6a21568f1149d61e2f54
>
> It is a 'plumbing' component which lets you use an automatically updated convexity value with the FuturesRateHelper, which you use to build a yield curve from futures prices. You would use it as the convexityAdjustment parameter to the FuturesRateHelper, constructor:
>
> https://www.quantlib.org/reference/class_quant_lib_1_1_futures_rate_he
> lper.html
>
> If you just want to calculate convexities, it would be simpler to call HullWhite::convexityBias directly.
>
> Either way, you need the same inputs:
>
> 1. the value date of the future
>
> 2. the maturity date of the future's notional deposit
>
> 3. the price of the future
>
> 4. the volatility of the yield curve at that point 
> (HullWhite::convexityBias calls this sigma); i am a bit hazy about the 
> exact nomenclature, but i believe this is a normal, aka Bachelier, 
> volatility
>
> 5. a mean-reversion constant (HullWhite::convexityBias calls this alpha, or a); i don't know if there's a way to derive this from market data, but i have been told it is done "empirically"
>
> Those are all simple parameters to HullWhite::convexityBias. If you want to use a FuturesConvAdjustmentQuote, you need to create one with handles linked to quotes, and set the values on the quotes.
>
> tom
>
> On Tue, 17 Jul 2018, Васильева Евгения Алексеевна wrote:
>
>> Dear All!
>>
>> First of all ,  my apologies for sending my question directly. I'm really have some technical issues with access to site.
>> Could you please help me, how to use  FuturesConvAdjustmentQuote in Quantlib to get convexity adjustment for 3M Eurodollar futures?
>>
>> Hope for your understanding!
>>
>> Thank you,
>> Evgenia
>
> --
> If you tolerate this, your children will be next.
> ----------------------------------------------------------------------
> -------- Check out the vibrant tech community on one of the world's 
> most engaging tech sites, Slashdot.org! http://sdm.link/slashdot 
> _______________________________________________
> QuantLib-users mailing list
> Qua...@li...<mailto:Qua...@li...
> eforge.net> 
> https://lists.sourceforge.net/lists/listinfo/quantlib-users
>
>

--
The future is still out there, somewhere.