quantlib-dev — Discussion of QuantLib development

[Quantlib-dev] Settlement days

[laaouini a. <laa...@ya...>]

Hello everybody,I have 2 questions about fixingDays parameter:+What are "fixing days" for an Euribor Index?+Why "settlement days" parameter  is equal to 2 for an Euribor object of C++ QuantLib?
---------------------------------------------------------<euribor.cpp>Euribor::Euribor(const Period& tenor,
                     const Handle<YieldTermStructure>& h)
    : IborIndex("Euribor", tenor,
                2, // settlement days
                EURCurrency(), TARGET(),
                euriborConvention(tenor), euriborEOM(tenor),
                Actual360(), h) {
        QL_REQUIRE(this->tenor().units()!=Days,
                   "for daily tenors (" << this->tenor() <<
                   ") dedicated DailyTenor constructor must be used");
    }

Re: [Quantlib-dev] QuantLibXL status

[Eric E. <eri...@re...>]

Hi Fabrice,

Many thanks for your understanding.  I am grateful for contributions to 
the old & new projects on the repos mentioned, on both of those repos I 
have received contributions which will be included in the next releases 
respectively.  I hope for the new build to supercede the old one, I set 
up regression tests to prove that the new build returns the same outputs 
as the old and so far those tests are ~80% successful.

Kind Regards,
Eric

On 2017-12-03 18:19, Fabrice Lecuyer wrote:
> Hi Eric,
>
> Many thanks for your quick reply.
> No need to apologize, I know all too well that real life as a tendency to
> get in the way and a big project like this one (let alone 2!) is no doubt
> very time consuming.
> I see indeed in your repositories that you have more advanced branches than
> master, I'll have a look at those.
> I don't have anything planned for now, but do you take in contributions for
> those projects?
> Also, do you foresee the new build to replace the old one, and ultimately do
> you plan to stop support of the old one?
>
> Regards,
> Fabrice
>
> Fabrice Lecuyer
> Quantitative Analyst
> fab...@qu...
> +33 6 63 71 19 71
>
> -----Original Message-----
> From: Eric Ehlers [mailto:eri...@re...]
> Sent: Sunday, December 3, 2017 6:02 PM
> To: Fabrice Lecuyer <fab...@qu...>
> Cc: qua...@li...
> Subject: Re: [Quantlib-dev] QuantLibXL status
>
> Hi Fabrice,
>
> Thanks for getting in touch, and apologies for the confusion.
>
> The old build is indeed overdue for a release.  I will have some time free
> to work on that next weekend and will try to get a release candidate out.
> Here is the repo for the old build:
>
> https://github.com/eehlers/QuantLibAddin-Old
>
> Work on the new build was about 80% complete then I had to put it on hold
> because of real life.  I will have a few days during the Christmas vacation
> to work on that and will try to push it over the finish line. Here is the
> documentation for compiling the new build from a git clone:
>
> http://quantlib.org/reposit/docs/latest/build_git_swig_windows.html
>
> That document refers to the master branches of each of the relevant repos
> but in fact the v1.8.x branches are more up to date.
>
> Kind Regards,
> Eric
>
> On 2017-12-03 17:04, Fabrice Lecuyer wrote:
>> Hi all,
>>
>> I’m a bit confused as to the status of QuantLibXL.
>>
>> The QuantLibXL, QuantLibAddin and ObjectHandler websites point to
>> https://sourceforge.net/projects/quantlib/files/ for the download of
>> the latest version (1.9.0) which seems to be lagging from QuantLib a
>> bit, with the last release made in May this year. If one wants to
>> compile the source code, it also seems to be available for all the
>> required modules.
>>
>> However, I’m at a loss to find the official source code repository.
>> The closest I could find is on Eric Ehlers GitHub and hasn’t been
>> updated for a couple of years. If one wants to contribute where does
>> one go?
>>
>> Secondly, there seems to have been an effort to create a new Excel
>> add-in using reposit this time:
>> http://quantlib.org/reposit/docs/latest/comparison.html. What is the
>> status of this effort? Are the official repositories the ones under
>> Eric’s GitHub ?
>>
>> Any clarification about the status of the 2 projects would be much
>> appreciated.
>>
>> Regards,
>>
>> *Fabrice Lecuyer*
>>
>> /Quantitative Analyst/
>>
>> fab...@qu... <mailto:fab...@qu...>
>>
>> +33 6 63 71 19 71
>>
>>
>>
>> ----------------------------------------------------------------------
>> -------- Check out the vibrant tech community on one of the world's
>> most engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>>
>>
>> _______________________________________________
>> QuantLib-dev mailing list
>> Qua...@li...
>> https://lists.sourceforge.net/lists/listinfo/quantlib-dev
>

Re: [Quantlib-dev] QuantLibXL status

[Eric E. <eri...@re...>]

Hi Fabrice,

Thanks for getting in touch, and apologies for the confusion.

The old build is indeed overdue for a release.  I will have some time 
free to work on that next weekend and will try to get a release 
candidate out.  Here is the repo for the old build:

https://github.com/eehlers/QuantLibAddin-Old

Work on the new build was about 80% complete then I had to put it on 
hold because of real life.  I will have a few days during the Christmas 
vacation to work on that and will try to push it over the finish line.  
Here is the documentation for compiling the new build from a git clone:

http://quantlib.org/reposit/docs/latest/build_git_swig_windows.html

That document refers to the master branches of each of the relevant 
repos but in fact the v1.8.x branches are more up to date.

Kind Regards,
Eric

On 2017-12-03 17:04, Fabrice Lecuyer wrote:
>
> Hi all,
>
> I’m a bit confused as to the status of QuantLibXL.
>
> The QuantLibXL, QuantLibAddin and ObjectHandler websites point to 
> https://sourceforge.net/projects/quantlib/files/ for the download of 
> the latest version (1.9.0) which seems to be lagging from QuantLib a 
> bit, with the last release made in May this year. If one wants to 
> compile the source code, it also seems to be available for all the 
> required modules.
>
> However, I’m at a loss to find the official source code repository. 
> The closest I could find is on Eric Ehlers GitHub and hasn’t been 
> updated for a couple of years. If one wants to contribute where does 
> one go?
>
> Secondly, there seems to have been an effort to create a new Excel 
> add-in using reposit this time: 
> http://quantlib.org/reposit/docs/latest/comparison.html. What is the 
> status of this effort? Are the official repositories the ones under 
> Eric’s GitHub ?
>
> Any clarification about the status of the 2 projects would be much 
> appreciated.
>
> Regards,
>
> *Fabrice Lecuyer*
>
> /Quantitative Analyst/
>
> fab...@qu... <mailto:fab...@qu...>
>
> +33 6 63 71 19 71
>
>
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>
>
> _______________________________________________
> QuantLib-dev mailing list
> Qua...@li...
> https://lists.sourceforge.net/lists/listinfo/quantlib-dev

Re: [Quantlib-dev] QuantLibXL status

[Fabrice L. <fab...@qu...>]

Hi Eric,

Many thanks for your quick reply.
No need to apologize, I know all too well that real life as a tendency to
get in the way and a big project like this one (let alone 2!) is no doubt
very time consuming.
I see indeed in your repositories that you have more advanced branches than
master, I'll have a look at those.
I don't have anything planned for now, but do you take in contributions for
those projects?
Also, do you foresee the new build to replace the old one, and ultimately do
you plan to stop support of the old one?

Regards,
Fabrice

Fabrice Lecuyer
Quantitative Analyst
fab...@qu...
+33 6 63 71 19 71

-----Original Message-----
From: Eric Ehlers [mailto:eri...@re...] 
Sent: Sunday, December 3, 2017 6:02 PM
To: Fabrice Lecuyer <fab...@qu...>
Cc: qua...@li...
Subject: Re: [Quantlib-dev] QuantLibXL status

Hi Fabrice,

Thanks for getting in touch, and apologies for the confusion.

The old build is indeed overdue for a release.  I will have some time free
to work on that next weekend and will try to get a release candidate out. 
Here is the repo for the old build:

https://github.com/eehlers/QuantLibAddin-Old

Work on the new build was about 80% complete then I had to put it on hold
because of real life.  I will have a few days during the Christmas vacation
to work on that and will try to push it over the finish line. Here is the
documentation for compiling the new build from a git clone:

http://quantlib.org/reposit/docs/latest/build_git_swig_windows.html

That document refers to the master branches of each of the relevant repos
but in fact the v1.8.x branches are more up to date.

Kind Regards,
Eric

On 2017-12-03 17:04, Fabrice Lecuyer wrote:
>
> Hi all,
>
> I’m a bit confused as to the status of QuantLibXL.
>
> The QuantLibXL, QuantLibAddin and ObjectHandler websites point to 
> https://sourceforge.net/projects/quantlib/files/ for the download of 
> the latest version (1.9.0) which seems to be lagging from QuantLib a 
> bit, with the last release made in May this year. If one wants to 
> compile the source code, it also seems to be available for all the 
> required modules.
>
> However, I’m at a loss to find the official source code repository. 
> The closest I could find is on Eric Ehlers GitHub and hasn’t been 
> updated for a couple of years. If one wants to contribute where does 
> one go?
>
> Secondly, there seems to have been an effort to create a new Excel 
> add-in using reposit this time:
> http://quantlib.org/reposit/docs/latest/comparison.html. What is the 
> status of this effort? Are the official repositories the ones under 
> Eric’s GitHub ?
>
> Any clarification about the status of the 2 projects would be much 
> appreciated.
>
> Regards,
>
> *Fabrice Lecuyer*
>
> /Quantitative Analyst/
>
> fab...@qu... <mailto:fab...@qu...>
>
> +33 6 63 71 19 71
>
>
>
> ----------------------------------------------------------------------
> -------- Check out the vibrant tech community on one of the world's 
> most engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>
>
> _______________________________________________
> QuantLib-dev mailing list
> Qua...@li...
> https://lists.sourceforge.net/lists/listinfo/quantlib-dev

[Quantlib-dev] QuantLibXL status

[Fabrice L. <fab...@qu...>]

Hi all,

 

I'm a bit confused as to the status of QuantLibXL.

The QuantLibXL, QuantLibAddin and ObjectHandler websites point to
https://sourceforge.net/projects/quantlib/files/ for the download of the
latest version (1.9.0) which seems to be lagging from QuantLib a bit, with
the last release made in May this year. If one wants to compile the source
code, it also seems to be available for all the required modules. 

However, I'm at a loss to find the official source code repository. The
closest I could find is on Eric Ehlers GitHub and hasn't been updated for a
couple of years. If one wants to contribute where does one go?

Secondly, there seems to have been an effort to create a new Excel add-in
using reposit this time:
http://quantlib.org/reposit/docs/latest/comparison.html. What is the status
of this effort? Are the official repositories the ones under Eric's GitHub ?

Any clarification about the status of the 2 projects would be much
appreciated.

 

Regards,

 

Fabrice Lecuyer

Quantitative Analyst

 <mailto:fab...@qu...> fab...@qu...

+33 6 63 71 19 71

 

Re: [Quantlib-dev] [Quantlib-users] How to construct a YieldTermStructure QuantLibObject from a list of Discount factors?

[laaouini a. <laa...@ya...>]

 Yes, I do it in the following way:
Date today(14, February, 2000);


boost::shared_ptr<IborIndex> makeIndex(std::vector<Date> dates, std::vector<DiscountFactor> discountfactors) { DayCounter dayCounter = Actual360();//Actual360(); RelinkableHandle<YieldTermStructure> termStructure; boost::shared_ptr<IborIndex> index(new Euribor6M(termStructure)); Date todaysDate = index->fixingCalendar().adjust(today); cout << "todaysDate " << todaysDate << endl; Settings::instance().evaluationDate() = todaysDate; for(int i=0;i<dates.size();i++) dates[i] = index->fixingCalendar().advance(dates[i],index->fixingDays(), Days); cout << "dates[0] " << dates[0] << endl; try {
     //new ZeroCurve(dates, rates, dayCounter);

 Calendar calendar = Calendar(); std::vector<Handle<Quote> > jumps = std::vector<Handle<Quote> >(); std::vector<Date> jumpDates = std::vector<Date>(); Linear interpolator = Linear(); Compounding compounding = Continuous; Frequency frequency = Daily;
 termStructure.linkTo(boost::shared_ptr<YieldTermStructure>(

 new InterpolatedDiscountCurve < LogLinear >( dates, discountfactors, dayCounter, calendar )
 )); } catch(Error e) { std::cout << "Error " << e.what()  << std::endl; } return index;}
boost::shared_ptr<IborIndex> makeIndex() {
 std::vector<Date> dates; dates.push_back(Date(14, February, 2000)); dates.push_back(Date(15, February, 2000)); dates.push_back(Date(15, August, 2000)); dates.push_back(Date(15, February, 2001)); dates.push_back(Date(15, August, 2001)); dates.push_back(Date(15, February, 2002)); dates.push_back(Date(15, August, 2002)); dates.push_back(Date(15, February, 2003)); dates.push_back(Date(15, August, 2003)); dates.push_back(Date(15, February, 2004)); dates.push_back(Date(15, August, 2004)); dates.push_back(Date(15, February, 2005)); dates.push_back(Date(15, August, 2005)); dates.push_back(Date(15, February, 2006)); dates.push_back(Date(15, August, 2006)); dates.push_back(Date(15, February, 2007)); dates.push_back(Date(15, August, 2007)); dates.push_back(Date(15, February, 2008)); dates.push_back(Date(15, August, 2008)); dates.push_back(Date(15, February, 2009)); dates.push_back(Date(15, August, 2009)); dates.push_back(Date(15, February, 2010)); dates.push_back(Date(15, August, 2010)); dates.push_back(Date(14, February, 2011)); dates.push_back(Date(14, February, 2012)); dates.push_back(Date(14, February, 2013)); dates.push_back(Date(14, February, 2014)); dates.push_back(Date(14, February, 2015));
 vector<DiscountFactor> discountFactors; discountFactors.push_back(1); discountFactors.push_back(0.999866337); discountFactors.push_back(0.97583485); discountFactors.push_back(0.952126374); discountFactors.push_back(0.929366524); discountFactors.push_back(0.906787023); discountFactors.push_back(0.885110975); discountFactors.push_back(0.863606688); discountFactors.push_back(0.842962833); discountFactors.push_back(0.82248256); discountFactors.push_back(0.802714439); discountFactors.push_back(0.783212024); discountFactors.push_back(0.764489942); discountFactors.push_back(0.745916213); discountFactors.push_back(0.728085659); discountFactors.push_back(0.710396393); discountFactors.push_back(0.693414913); discountFactors.push_back(0.676567994); discountFactors.push_back(0.660306885); discountFactors.push_back(0.644264345); discountFactors.push_back(0.6288637); discountFactors.push_back(0.61358509); discountFactors.push_back(0.59891781); discountFactors.push_back(0.584398104); discountFactors.push_back(0.556480797); discountFactors.push_back(0.52982607); discountFactors.push_back(0.504515727); discountFactors.push_back(0.480414485);


 return makeIndex(dates, discountFactors);
}
    Le vendredi 1 décembre 2017 à 15:30:50 UTC+1, <fab...@gm...> a écrit :  
 
 #yiv3461759544 #yiv3461759544 -- _filtered #yiv3461759544 {font-family:Helvetica;panose-1:2 11 6 4 2 2 2 2 2 4;} _filtered #yiv3461759544 {panose-1:2 4 5 3 5 4 6 3 2 4;} _filtered #yiv3461759544 {font-family:Calibri;panose-1:2 15 5 2 2 2 4 3 2 4;}#yiv3461759544 #yiv3461759544 p.yiv3461759544MsoNormal, #yiv3461759544 li.yiv3461759544MsoNormal, #yiv3461759544 div.yiv3461759544MsoNormal {margin:0in;margin-bottom:.0001pt;font-size:11.0pt;}#yiv3461759544 a:link, #yiv3461759544 span.yiv3461759544MsoHyperlink {color:#0563C1;text-decoration:underline;}#yiv3461759544 a:visited, #yiv3461759544 span.yiv3461759544MsoHyperlinkFollowed {color:#954F72;text-decoration:underline;}#yiv3461759544 p.yiv3461759544msonormal0, #yiv3461759544 li.yiv3461759544msonormal0, #yiv3461759544 div.yiv3461759544msonormal0 {margin-right:0in;margin-left:0in;font-size:11.0pt;}#yiv3461759544 span.yiv3461759544EmailStyle18 {color:windowtext;}#yiv3461759544 .yiv3461759544MsoChpDefault {font-size:10.0pt;} _filtered #yiv3461759544 {margin:70.85pt 70.85pt 70.85pt 70.85pt;}#yiv3461759544 div.yiv3461759544WordSection1 {}#yiv3461759544 
Hi,

  

I believe you can create such a curve using one of the 3 public constructors of the InterpolatedDiscountCurve class (ql\termstructures\yield\discountcurve.hpp), which is documented here: http://quantlib.org/reference/class_quant_lib_1_1_interpolated_discount_curve.html#details

  

Regards,

Fabrice Lecuyer

  

From: laaouini anas [mailto:laa...@ya...] 
Sent: Thursday, November 30, 2017 4:35 PM
To: lui...@gm...; qua...@li...; qua...@li...
Subject: [Quantlib-users] How to construct a YieldTermStructure QuantLibObject from a list of Discount factors?

  

Hello everybody,

  

I have a question for you:

How to construct a YieldTermStructure C++ QuantLib Object from a list of Discount factors?

The Quantlib classes and methods are not sufficiently documented. For example, if someone wants

to know how a given method of a class works, he must look at its implementation. 
  

Re: [Quantlib-dev] [Quantlib-users] How to construct a YieldTermStructure QuantLibObject from a list of Discount factors?

[<fab...@gm...>]

Hi,

 

I believe you can create such a curve using one of the 3 public constructors of the InterpolatedDiscountCurve class (ql\termstructures\yield\discountcurve.hpp), which is documented here: http://quantlib.org/reference/class_quant_lib_1_1_interpolated_discount_curve.html#details

 

Regards,

Fabrice Lecuyer

 

From: laaouini anas [mailto:laa...@ya...] 
Sent: Thursday, November 30, 2017 4:35 PM
To: lui...@gm...; qua...@li...; qua...@li...
Subject: [Quantlib-users] How to construct a YieldTermStructure QuantLibObject from a list of Discount factors?

 

Hello everybody,

 

I have a question for you:

How to construct a YieldTermStructure C++ QuantLib Object from a list of Discount factors?

The Quantlib classes and methods are not sufficiently documented. For example, if someone wants

to know how a given method of a class works, he must look at its implementation. 

Re: [Quantlib-dev] [Quantlib-users] How to construct a YieldTermStructure QuantLibObject from a list of Discount factors?

[Liu Y. <yml...@gm...>]

I think you can refer to the tutorial provided on documentation link: http://quantlib.org/slides/dima-ql-intro-2.pdf#page87

There you can find the example to create a yield curve from discount factors input.

Yiming

> On Nov 30, 2017, at 10:35 AM, laaouini anas via QuantLib-users <qua...@li...> wrote:
> 
> Hello everybody,
> 
> I have a question for you:
> How to construct a YieldTermStructure C++ QuantLib Object from a list of Discount factors?
> The Quantlib classes and methods are not sufficiently documented. For example, if someone wants
> to know how a given method of a class works, he must look at its implementation. 
> ------------------------------------------------------------------------------
> 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...
> https://lists.sourceforge.net/lists/listinfo/quantlib-users

[Quantlib-dev] How to construct a YieldTermStructure QuantLibObject from a list of Discount factors?

[laaouini a. <laa...@ya...>]

Hello everybody,
I have a question for you:How to construct a YieldTermStructure C++ QuantLib Object from a list of Discount factors?
The Quantlib classes and methods are not sufficiently documented. For example, if someone wantsto know how a given method of a class works, he must look at its implementation. 

[Quantlib-dev] Quntlib install problem- Linux- ubuntu 16.04

[wassim <jem...@ou...>]

I have been using quantlib on a windows Python Version and it works
perfectly. 

I’am trying to get it t work on my linux version but I have been stuck for a
good while on compiling the files with boost. You will find below the error
message.

 
*********************
quantlib_wrap.cpp:2346:27: fatal error: ql/quantlib.hpp: No such file or
directory
compilation terminated.
Makefile.quantlib:336: recipe for target 'quantlib_wrap.o' failed
make[3]: *** [quantlib_wrap.o] Error 1
make[3]: Leaving directory '/root/djangov/FastNet/QuantLib-SWIG-1.7/Perl'
Makefile:426: recipe for target '.build-stamp' failed
make[2]: *** [.build-stamp] Error 2
make[2]: Leaving directory '/root/djangov/FastNet/QuantLib-SWIG-1.7/Perl'
Makefile:220: recipe for target 'all' failed
make[1]: *** [all] Error 2
make[1]: Leaving directory '/root/djangov/FastNet/QuantLib-SWIG-1.7/Perl'
Makefile:349: recipe for target 'all-recursive' failed
make: *** [all-recursive] Error 1

*********************


I would more that grateful if anybody can direct me to where I can find help



--
Sent from: http://quantlib.10058.n7.nabble.com/quantlib-dev-f8818.html

[Quantlib-dev] Pricing a Bermudan Swaption under LFM model -LongStaff Schwartz method

[laaouini a. <laa...@ya...>]

Hello,How can we price a bermudan swaption under the forward Libor Model, with QuantLib, using LongStaff and Schwartz method?

Re: [Quantlib-dev] [Quantlib-users] dates() accessor in YieldTermStructureHandle

[Luigi B. <lui...@gm...>]

Hello Fabrice,
    we had the same problem with other hierarchies, and we exported a few
utility functions to perform the casts.  See for instance as_coupon and
similar in cashflows.i, or as_iborindex in indexes.i.  You might export
something similar for curves.

Luigi


On Thu, Nov 23, 2017 at 11:38 AM Fabrice Lecuyer <
fab...@qu...> wrote:

> Hi all,
>
>
>
> I’ve run into an “inheritance issue” for which I’m uncertain of the way
> forward. Note that I’m using QuantLib in C# through SWIG.
>
>
>
> I’m building a set of yield curves (PiecewiseYieldCurve) of various
> flavors: some are inheriting InterpolatedDiscountCurve, some from
> InterpolatedZeroCurve, etc.
>
> In order to use those curves  as exogenous discounting curves for other
> calibration instruments (SwapRateHelper for instance) and to be able to
> store them all in a single place, I store them in a Dictionary<string,
> YieldTermStructureHandle>.
>
> I also want to be able to display them using dates, not times, as the
> pillar points. While I can call discounts() on YieldTermStructureHandle,
> the dates() accessor is unavailable, because it is not defined in
> YieldTermStructure. It doesn’t seem that I can dereference the handle
> (I’m in C#) and downcast the YieldTermStructure to a PiecewiseYieldCurve that
> I could query for dates.
>
> Furthermore, in C++ the dates() accessor is never defined as a virtual
> function of a base class, despite being present in all the classes
> inheriting from InterpolatedCurve<Interpolator>. I realize this would not
> solve my issue, but maybe a DateInterpolatedCurve class, between
> InterpolatedCurve and say InterpolatedDiscountCurve, which would hold
> mutable std::vector<Date> dates_ and const std::vector<Date>& dates()
> const; would make sense.
>
>
>
> What’s the best way forward in order for me to be able to retrieve the
> pillar dates of those curves while retaining the capacity to store them all
> under one base class handle type regardless of their “flavor”?
>
>
>
> Regards,
>
> Fabrice Lecuyer
>
>
>
> ------------------------------------------------------------------------------
> 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...
> https://lists.sourceforge.net/lists/listinfo/quantlib-users
>

[Quantlib-dev] dates() accessor in YieldTermStructureHandle

[Fabrice L. <fab...@qu...>]

Hi all,

 

I've run into an "inheritance issue" for which I'm uncertain of the way
forward. Note that I'm using QuantLib in C# through SWIG.

 

I'm building a set of yield curves (PiecewiseYieldCurve) of various flavors:
some are inheriting InterpolatedDiscountCurve, some from
InterpolatedZeroCurve, etc.

In order to use those curves  as exogenous discounting curves for other
calibration instruments (SwapRateHelper for instance) and to be able to
store them all in a single place, I store them in a Dictionary<string,
YieldTermStructureHandle>. 

I also want to be able to display them using dates, not times, as the pillar
points. While I can call discounts() on YieldTermStructureHandle, the
dates() accessor is unavailable, because it is not defined in
YieldTermStructure. It doesn't seem that I can dereference the handle (I'm
in C#) and downcast the YieldTermStructure to a PiecewiseYieldCurve that I
could query for dates.

Furthermore, in C++ the dates() accessor is never defined as a virtual
function of a base class, despite being present in all the classes
inheriting from InterpolatedCurve<Interpolator>. I realize this would not
solve my issue, but maybe a DateInterpolatedCurve class, between
InterpolatedCurve and say InterpolatedDiscountCurve, which would hold
mutable std::vector<Date> dates_ and const std::vector<Date>& dates() const;
would make sense.

 

What's the best way forward in order for me to be able to retrieve the
pillar dates of those curves while retaining the capacity to store them all
under one base class handle type regardless of their "flavor"?

 

Regards,

Fabrice Lecuyer

 

Re: [Quantlib-dev] Difference curves

[Luigi B. <lui...@gm...>]

Yes, they should probably have the same reference date.  The same day
counter would make the implementation easier, but you can probably make it
work even with different ones.

Luigi


On Sun, Nov 12, 2017 at 8:48 AM Francois Botha <ig...@gm...> wrote:

> Luigi,
>
> Do you agree this approach makes sense only if both curves have the same
> reference date? Or can you think of valid cases where that's not the case?
>
> Thanks
> Francois
>
>
> On 31 Oct 2017 4:07 p.m., "Luigi Ballabio" <lui...@gm...>
> wrote:
>
> It could add value, but then again, I haven't seen many uses of
> CompositeQuote these past few years :)
> If all you need is the difference, I'd start with that---unless you
> estimate that the additional work to make it generic is not much.
>
> Later,
>     Luigi
>
>
> On Mon, Oct 30, 2017 at 4:54 PM Francois Botha <ig...@gm...> wrote:
>
>> It will be an interest-rate term structure. I will generate my own
>> inflation term structure from this. There could be other,
>> non-inflation-related examples too where this could be helpful.
>>
>> In fact, the CompositeQuote (for Quotes) is a good analogy of what I
>> would like to achieve, so maybe a CompositeYieldTermStructure is a good
>> class name? The user would be able to provide a BinaryFunction which
>> operates on the zero rates of the 2 inputs curves. Does that sound like it
>> could add value?
>>
>> thanks
>>
>> Francois Botha
>>
>> On 30 October 2017 at 13:19, Luigi Ballabio <lui...@gm...>
>> wrote:
>>
>>> Hi Francois,
>>>     there's no such curve at this time.  Question: should this be an
>>> interest-rate term structure or an inflation term structure?
>>>
>>> Luigi
>>>
>>>
>>> On Mon, Oct 30, 2017 at 10:29 AM Francois Botha <ig...@gm...>
>>> wrote:
>>>
>>>> Hi,
>>>>
>>>> I have two YieldTermStructures representing a nominal and real yield
>>>> curve and would like to generate a new YieldTermStructure based on the
>>>> difference of zero rates (NACC) between them to represent the market
>>>> expectation of inflation.
>>>>
>>>> How would I calculate this currently? I can't find a YieldTermStructure
>>>> subclass that does this. Alternatively, if I can find a YieldTermStructure
>>>> that "negates" another YieldTermStructure, I would be able to use it as a
>>>> spread.
>>>>
>>>> If this isn't currently possible, I'd be happy to submit a PR, but I
>>>> would like some help on what to call these new classes.
>>>>
>>>> thanks
>>>> Francois Botha
>>>>
>>>> ------------------------------------------------------------------------------
>>>> Check out the vibrant tech community on one of the world's most
>>>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>>>> _______________________________________________
>>>> QuantLib-dev mailing list
>>>> Qua...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/quantlib-dev
>>>>
>>>
>>
>

Re: [Quantlib-dev] Difference curves

[Francois B. <ig...@gm...>]

Luigi,

Do you agree this approach makes sense only if both curves have the same
reference date? Or can you think of valid cases where that's not the case?

Thanks
Francois

On 31 Oct 2017 4:07 p.m., "Luigi Ballabio" <lui...@gm...> wrote:

It could add value, but then again, I haven't seen many uses of
CompositeQuote these past few years :)
If all you need is the difference, I'd start with that---unless you
estimate that the additional work to make it generic is not much.

Later,
    Luigi


On Mon, Oct 30, 2017 at 4:54 PM Francois Botha <ig...@gm...> wrote:

> It will be an interest-rate term structure. I will generate my own
> inflation term structure from this. There could be other,
> non-inflation-related examples too where this could be helpful.
>
> In fact, the CompositeQuote (for Quotes) is a good analogy of what I would
> like to achieve, so maybe a CompositeYieldTermStructure is a good class
> name? The user would be able to provide a BinaryFunction which operates on
> the zero rates of the 2 inputs curves. Does that sound like it could add
> value?
>
> thanks
>
> Francois Botha
>
> On 30 October 2017 at 13:19, Luigi Ballabio <lui...@gm...>
> wrote:
>
>> Hi Francois,
>>     there's no such curve at this time.  Question: should this be an
>> interest-rate term structure or an inflation term structure?
>>
>> Luigi
>>
>>
>> On Mon, Oct 30, 2017 at 10:29 AM Francois Botha <ig...@gm...> wrote:
>>
>>> Hi,
>>>
>>> I have two YieldTermStructures representing a nominal and real yield
>>> curve and would like to generate a new YieldTermStructure based on the
>>> difference of zero rates (NACC) between them to represent the market
>>> expectation of inflation.
>>>
>>> How would I calculate this currently? I can't find a YieldTermStructure
>>> subclass that does this. Alternatively, if I can find a YieldTermStructure
>>> that "negates" another YieldTermStructure, I would be able to use it as a
>>> spread.
>>>
>>> If this isn't currently possible, I'd be happy to submit a PR, but I
>>> would like some help on what to call these new classes.
>>>
>>> thanks
>>> Francois Botha
>>> ------------------------------------------------------------
>>> ------------------
>>> Check out the vibrant tech community on one of the world's most
>>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot______
>>> _________________________________________
>>> QuantLib-dev mailing list
>>> Qua...@li...
>>> https://lists.sourceforge.net/lists/listinfo/quantlib-dev
>>>
>>
>

Re: [Quantlib-dev] Difference curves

[Luigi B. <lui...@gm...>]

It could add value, but then again, I haven't seen many uses of
CompositeQuote these past few years :)
If all you need is the difference, I'd start with that---unless you
estimate that the additional work to make it generic is not much.

Later,
    Luigi


On Mon, Oct 30, 2017 at 4:54 PM Francois Botha <ig...@gm...> wrote:

> It will be an interest-rate term structure. I will generate my own
> inflation term structure from this. There could be other,
> non-inflation-related examples too where this could be helpful.
>
> In fact, the CompositeQuote (for Quotes) is a good analogy of what I would
> like to achieve, so maybe a CompositeYieldTermStructure is a good class
> name? The user would be able to provide a BinaryFunction which operates on
> the zero rates of the 2 inputs curves. Does that sound like it could add
> value?
>
> thanks
>
> Francois Botha
>
> On 30 October 2017 at 13:19, Luigi Ballabio <lui...@gm...>
> wrote:
>
>> Hi Francois,
>>     there's no such curve at this time.  Question: should this be an
>> interest-rate term structure or an inflation term structure?
>>
>> Luigi
>>
>>
>> On Mon, Oct 30, 2017 at 10:29 AM Francois Botha <ig...@gm...> wrote:
>>
>>> Hi,
>>>
>>> I have two YieldTermStructures representing a nominal and real yield
>>> curve and would like to generate a new YieldTermStructure based on the
>>> difference of zero rates (NACC) between them to represent the market
>>> expectation of inflation.
>>>
>>> How would I calculate this currently? I can't find a YieldTermStructure
>>> subclass that does this. Alternatively, if I can find a YieldTermStructure
>>> that "negates" another YieldTermStructure, I would be able to use it as a
>>> spread.
>>>
>>> If this isn't currently possible, I'd be happy to submit a PR, but I
>>> would like some help on what to call these new classes.
>>>
>>> thanks
>>> Francois Botha
>>>
>>> ------------------------------------------------------------------------------
>>> Check out the vibrant tech community on one of the world's most
>>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>>> _______________________________________________
>>> QuantLib-dev mailing list
>>> Qua...@li...
>>> https://lists.sourceforge.net/lists/listinfo/quantlib-dev
>>>
>>
>

Re: [Quantlib-dev] Difference curves

[Francois B. <ig...@gm...>]

It will be an interest-rate term structure. I will generate my own
inflation term structure from this. There could be other,
non-inflation-related examples too where this could be helpful.

In fact, the CompositeQuote (for Quotes) is a good analogy of what I would
like to achieve, so maybe a CompositeYieldTermStructure is a good class
name? The user would be able to provide a BinaryFunction which operates on
the zero rates of the 2 inputs curves. Does that sound like it could add
value?

thanks

Francois Botha

On 30 October 2017 at 13:19, Luigi Ballabio <lui...@gm...>
wrote:

> Hi Francois,
>     there's no such curve at this time.  Question: should this be an
> interest-rate term structure or an inflation term structure?
>
> Luigi
>
>
> On Mon, Oct 30, 2017 at 10:29 AM Francois Botha <ig...@gm...> wrote:
>
>> Hi,
>>
>> I have two YieldTermStructures representing a nominal and real yield
>> curve and would like to generate a new YieldTermStructure based on the
>> difference of zero rates (NACC) between them to represent the market
>> expectation of inflation.
>>
>> How would I calculate this currently? I can't find a YieldTermStructure
>> subclass that does this. Alternatively, if I can find a YieldTermStructure
>> that "negates" another YieldTermStructure, I would be able to use it as a
>> spread.
>>
>> If this isn't currently possible, I'd be happy to submit a PR, but I
>> would like some help on what to call these new classes.
>>
>> thanks
>> Francois Botha
>> ------------------------------------------------------------
>> ------------------
>> Check out the vibrant tech community on one of the world's most
>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot______
>> _________________________________________
>> QuantLib-dev mailing list
>> Qua...@li...
>> https://lists.sourceforge.net/lists/listinfo/quantlib-dev
>>
>

Re: [Quantlib-dev] Question about Schedule dates calculation using Forward rule

[Luigi B. <lui...@gm...>]

Hello Simon,
    I'm guessing that Dimitri's slides contain the output of the library at
the time he wrote them.  The schedule code has changed since, and now you
get the new output.  I'm not sure which result is "correct"; you're telling
the schedule to enforce end of months and passing an end date in the middle
of the month, which might be a bit of a corner case.  In any case, you can
select the behavior by changing the termination date convention; by passing
Unadjusted, you'll get Dimitri's output.

Luigi


On Mon, Oct 30, 2017 at 10:14 AM Simon Courtenage <cou...@gm...>
wrote:

> Hi all,
>
> I was reading Dimitri Reiswich's slides on Quantlib (found at
> http://quantlib.org/slides/dima-ql-intro-1.pdf), and in particular, the
> slides about the Schedule class. On slide 36, there is some example code to
> show how to construct a Schedule.  This code is below:
>
> Date  begin(30,September ,2009) ,  end(15,Jun ,2012);
> Calendar  myCal=Japan ();
>
> BusinessDayConvention  bdC=BusinessDayConvention(Following );
> Period  myTenor(6,Months );
> DateGeneration ::Rule  myRule=DateGeneration :: Forward;
>
> Schedule  mySched(begin ,end ,myTenor ,myCal ,bdC ,bdC ,myRule ,true);
>
> std::vector <Date > finalSched=mySched.dates ();
> BOOST_FOREACH(Date d,finalSched) std::cout << d << std::endl;
>
> On slide 37 is its example output, which is:
>
> September 30th, 2009
> March 31st, 2010
> September 30th, 2010
> March 31st, 2011
> September 30th, 2011
> March 30th, 2012
> June 15th, 2012
>
> When I compiled and ran this code myself, I got
>
> June 15th, 2012
>
> September 30th, 2009
>
> March 31st, 2010
>
> September 30th, 2010
>
> March 31st, 2011
>
> September 30th, 2011
>
> March 30th, 2012
>
> June 29th, 2012
> The difference being the last date - Jun 29th instead of Jun 15th.  The last date is later than the termination date provided in the constructor call.
>
> My question is which behaviour is expected and correct?  June 15th (as in Dimitri's slides) or June 29th, the actual result?
>
> If Dimitri's version is correct, then I suspect that this code after the case statements for date calculation in the Schedule constructor (dealing with effective dates) will need to change, since its the code responsible for changing June 15th t June 29th in the example above.
>
> if (terminationDateConvention != Unadjusted) {
>       dates_.front() = calendar_.endOfMonth(dates_.front());
>       dates_.back() = calendar_.endOfMonth(dates_.back());
> }
>
> I'm compiling against the latest QL code cloned from github yesterday.
>
> Thanks
>
> Simon
>
>
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> QuantLib-dev mailing list
> Qua...@li...
> https://lists.sourceforge.net/lists/listinfo/quantlib-dev
>

Re: [Quantlib-dev] Difference curves

[Luigi B. <lui...@gm...>]

Hi Francois,
    there's no such curve at this time.  Question: should this be an
interest-rate term structure or an inflation term structure?

Luigi


On Mon, Oct 30, 2017 at 10:29 AM Francois Botha <ig...@gm...> wrote:

> Hi,
>
> I have two YieldTermStructures representing a nominal and real yield curve
> and would like to generate a new YieldTermStructure based on the difference
> of zero rates (NACC) between them to represent the market expectation of
> inflation.
>
> How would I calculate this currently? I can't find a YieldTermStructure
> subclass that does this. Alternatively, if I can find a YieldTermStructure
> that "negates" another YieldTermStructure, I would be able to use it as a
> spread.
>
> If this isn't currently possible, I'd be happy to submit a PR, but I would
> like some help on what to call these new classes.
>
> thanks
> Francois Botha
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> QuantLib-dev mailing list
> Qua...@li...
> https://lists.sourceforge.net/lists/listinfo/quantlib-dev
>

[Quantlib-dev] Difference curves

[Francois B. <ig...@gm...>]

Hi,

I have two YieldTermStructures representing a nominal and real yield curve
and would like to generate a new YieldTermStructure based on the difference
of zero rates (NACC) between them to represent the market expectation of
inflation.

How would I calculate this currently? I can't find a YieldTermStructure
subclass that does this. Alternatively, if I can find a YieldTermStructure
that "negates" another YieldTermStructure, I would be able to use it as a
spread.

If this isn't currently possible, I'd be happy to submit a PR, but I would
like some help on what to call these new classes.

thanks
Francois Botha

[Quantlib-dev] Question about Schedule dates calculation using Forward rule

[Simon C. <cou...@gm...>]

Hi all,

I was reading Dimitri Reiswich's slides on Quantlib (found at
http://quantlib.org/slides/dima-ql-intro-1.pdf), and in particular, the
slides about the Schedule class. On slide 36, there is some example code to
show how to construct a Schedule.  This code is below:

Date  begin(30,September ,2009) ,  end(15,Jun ,2012);
Calendar  myCal=Japan ();

BusinessDayConvention  bdC=BusinessDayConvention(Following );
Period  myTenor(6,Months );
DateGeneration ::Rule  myRule=DateGeneration :: Forward;

Schedule  mySched(begin ,end ,myTenor ,myCal ,bdC ,bdC ,myRule ,true);

std::vector <Date > finalSched=mySched.dates ();
BOOST_FOREACH(Date d,finalSched) std::cout << d << std::endl;

On slide 37 is its example output, which is:

September 30th, 2009
March 31st, 2010
September 30th, 2010
March 31st, 2011
September 30th, 2011
March 30th, 2012
June 15th, 2012

When I compiled and ran this code myself, I got

June 15th, 2012

September 30th, 2009

March 31st, 2010

September 30th, 2010

March 31st, 2011

September 30th, 2011

March 30th, 2012

June 29th, 2012
The difference being the last date - Jun 29th instead of Jun 15th.
The last date is later than the termination date provided in the
constructor call.

My question is which behaviour is expected and correct?  June 15th (as
in Dimitri's slides) or June 29th, the actual result?

If Dimitri's version is correct, then I suspect that this code after
the case statements for date calculation in the Schedule constructor
(dealing with effective dates) will need to change, since its the code
responsible for changing June 15th t June 29th in the example above.

if (terminationDateConvention != Unadjusted) {
      dates_.front() = calendar_.endOfMonth(dates_.front());
      dates_.back() = calendar_.endOfMonth(dates_.back());
}

I'm compiling against the latest QL code cloned from github yesterday.

Thanks

Simon

Re: [Quantlib-dev] FW: QuantLib User Meeting 2017 at IKB (Duesseldorf)

[Luigi B. <lui...@gm...>]

The form is at <https://goo.gl/forms/ZYlpjdgSK7uyrKp73>.

Luigi


On Thu, Oct 12, 2017 at 1:43 PM Ioannis Rigopoulos <qua...@de...>
wrote:

> Hi Michael,
>
> What is the application procedure for anyone interested in participating
> as a speaker in that or a similar forthcoming event?
>
> I could not find a related link in your post.
>
> Regards
>
> Ioannis
>
>
> On 11.10.2017 14:29, Michael von den Driesch wrote:
> > Dear all,
> >
> > sorry for the html Format in my first mail.
> >
> > This should work now.
> >
> > Click here for details: http://quantlib.org/slides/qlum17/flyer.pdf
> > You can also directly go to registration here:
> https://goo.gl/forms/ZYlpjdgSK7uyrKp73
> >
> > Best
> >
> > Michael
> >
> >
> >
> ------------------------------------------------------------------------------
> > Check out the vibrant tech community on one of the world's most
> > engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> > _______________________________________________
> > QuantLib-dev mailing list
> > Qua...@li...
> > https://lists.sourceforge.net/lists/listinfo/quantlib-dev
>
>
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> QuantLib-dev mailing list
> Qua...@li...
> https://lists.sourceforge.net/lists/listinfo/quantlib-dev
>

Re: [Quantlib-dev] FW: QuantLib User Meeting 2017 at IKB (Duesseldorf)

[Ioannis R. <qua...@de...>]

Hi Michael,

What is the application procedure for anyone interested in participating 
as a speaker in that or a similar forthcoming event?

I could not find a related link in your post.

Regards

Ioannis


On 11.10.2017 14:29, Michael von den Driesch wrote:
> Dear all,
>
> sorry for the html Format in my first mail.
>
> This should work now.
>
> Click here for details: http://quantlib.org/slides/qlum17/flyer.pdf
> You can also directly go to registration here: https://goo.gl/forms/ZYlpjdgSK7uyrKp73
>
> Best
>
> Michael
>   
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> QuantLib-dev mailing list
> Qua...@li...
> https://lists.sourceforge.net/lists/listinfo/quantlib-dev

[Quantlib-dev] FW: QuantLib User Meeting 2017 at IKB (Duesseldorf)

[Michael v. d. D. <mdr...@we...>]

Dear all,

sorry for the html Format in my first mail.

This should work now.

Click here for details: http://quantlib.org/slides/qlum17/flyer.pdf
You can also directly go to registration here: https://goo.gl/forms/ZYlpjdgSK7uyrKp73

Best

Michael 
 

[Quantlib-dev] QuantLib User Meeting 2017 at IKB (Duesseldorf)

[Michael v. d. D. <mdr...@we...>]

Dear all,
 
registration is open for the 2017 QuantLib User Meeting at IKB in Düsseldorf.
 
Click here for details: http://quantlib.org/slides/qlum17/flyer.pdf
You can also directly go to registration here: https://goo.gl/forms/ZYlpjdgSK7uyrKp73[https://goo.gl/forms/ZYlpjdgSK7uyrKp73]
 
I'm looking forward to see you at the event.
 
Best
 
Michael