quantlib-dev — Discussion of QuantLib development

[Quantlib-dev] [ quantlib-Bugs-3466468 ] SWIG/CSharp/VS2010

[SourceForge.net <no...@so...>]

Bugs item #3466468, was opened at 2011-12-28 04:03
Message generated for change (Tracker Item Submitted) made by 
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=112740&aid=3466468&group_id=12740

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: https://www.google.com/accounts ()
Assigned to: Nobody/Anonymous (nobody)
Summary: SWIG/CSharp/VS2010

Initial Comment:
Hello I have been using QuantLib in Visual Studio 2010, and have come across the following issues:

1) There is an %include in the SWIG wrapper which #defines SWIGSTDCALL to nothing if the SWIG target is CSharp. This causes P/Invoke to throw an exception under the .NET 4.0 Framework. (see http://msdn.microsoft.com/en-us/library/ee941656(v=VS.100).aspx, "Platform Invoke") because the wrapper functions are not __stdcall.

2) In order to use the numerical integration classes from C# I had to add the following code to the SWIG wrapper, perhaps you could consider integrating it into the next release:

=== types.i ===
...
#elif defined(SWIGCSHARP)
// The following macro defines a typemap between boost::function<RTYPE(PTYPEA)> and a corresponding .NET delegate type
%define %cs_func1(RTYPE, friendlyRTYPE, PTYPEA, friendlyPTYPEA) 
	#define BFUNC boost::function<RTYPE(PTYPEA)>
	#define DELEGATE friendlyPTYPEA##To##friendlyRTYPE##Delegate

	%pragma(csharp) modulecode=%{
		public delegate RTYPE friendlyPTYPEA##To##friendlyRTYPE##Delegate(PTYPEA a);
	%}

        %typemap(ctype) BFUNC, BFUNC& "void*" 
        %typemap(in) BFUNC { $1 = (RTYPE (__stdcall *)(PTYPEA)) $input; }
        %typemap(in) BFUNC& { $1 = (RTYPE (__stdcall *)(PTYPEA)) $input; }
        %typemap(imtype, out="IntPtr", noblock=1) BFUNC, BFUNC& { NQuantLibc.DELEGATE }
        %typemap(cstype, out="IntPtr", noblock=1) BFUNC, BFUNC& { NQuantLibc.DELEGATE }
        %typemap(csin, noblock=1) BFUNC, BFUNC& { $csinput }

	#undef DELEGATE
	#undef BFUNC
%enddef
// Typemap boost::function<double(double)> to NQuantLibc.DoubleToDoubleDelegate
%cs_func1(double, Double, double, Double)
#endif
...


=== integrals.i ===
...
%define INTEGRATION_METHODS
...
#elif defined(SWIGCSHARP)
Real Call(boost::function<double(double)> f, Real a, Real b)
{
	return (*self)(f, a, b);
}
#endif

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=112740&aid=3466468&group_id=12740

Re: [Quantlib-dev] Re petitive function calls in discountingbondengine.cpp

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

Done, thanks.

    Luigi


On Thu, Dec 22, 2011 at 11:46 PM, StephenWong <ste...@gm...> wrote:
>
> Within the file ql/pricingengines/bond/discountingbondengine.cpp there is the
> DiscountingBondEngine::calculate() method. Within this method, it sets
> results_.value and results_.settlementValue by calling the
> CashFlows::npv(...) function each time. It does this even if all the
> arguments of the two calls are identical.
>
> Can we not check the arguments first and if they are identical, just call
> the CashFlows::npv(...) once? It is a bit of an eye sore for me to see all
> the other functions being called within CashFlows::npv(...) over and over
> even if there is no such need?
>
> More explicitly, after the line
>
>        results_.value = CashFlows::npv(arguments_.cashflows,
>                                        **discountCurve_,
>                                        includeRefDateFlows,
>                                        results_.valuationDate,
>                                        results_.valuationDate);
>
> just insert
>
> if ( includeRefDateFlows == false && results_.valuationDate ==
> arguments_.settlementDate )
>     results_.settlementValue = results.value;
> else {
>
>     results_.settlementValue = CashFlows::npv(arguments_.cashflows,
>
> **discountCurve_,
>                                                                false,
>
> arguments_.settlementDate,
>
> arguments_.settlementDate);
>
> }
>
> I know this does not look like much but it is a bit of a pain in the neck if
> one is going through the code, debugging etc and see all the function calls
> being made over and over when there is no need.
>
>
>
> --
> View this message in context: http://old.nabble.com/Repetitive-function-calls-in-discountingbondengine.cpp-tp33026149p33026149.html
> Sent from the quantlib-dev mailing list archive at Nabble.com.
>
>
> ------------------------------------------------------------------------------
> Write once. Port to many.
> Get the SDK and tools to simplify cross-platform app development. Create
> new or port existing apps to sell to consumers worldwide. Explore the
> Intel AppUpSM program developer opportunity. appdeveloper.intel.com/join
> http://p.sf.net/sfu/intel-appdev
> _______________________________________________
> QuantLib-dev mailing list
> Qua...@li...
> https://lists.sourceforge.net/lists/listinfo/quantlib-dev

Re: [Quantlib-dev] coerce problem under Linux/amd64-gcc

[Lars C. <lar...@gm...>]

Hello,

this problem is not restricted to SimpleQuote/FlatForward. Under Linux I 
see for example the following error messages:
INFO - ERROR: qlGeneralizedBlackScholesProcess: Error retrieving object 
with id 'blackvol1' - unable to convert reference to type 
'N13QuantLibAddin21BlackVolTermStructureE' found instead 
'N13QuantLibAddin16BlackConstantVolE'
INFO - ERROR: qlVanillaOption: Error retrieving object with id 
'eu_payoff' - unable to convert reference to type 
'N13QuantLibAddin17StrikedTypePayoffE' found instead 
'N13QuantLibAddin17StrikedTypePayoffE'
The second error does not make sense, does it?

Regards, 
   Lars 


-------- Original-Nachricht --------
Datum: Fri, 23 Dec 2011 21:50:06 +0100
Von: "Lars Callenbach" <lar...@gm...>
An: qua...@li...
Betreff: coerce problem under Linux/amd64-gcc

Hello,

compiling 1.1.0 with calc addins under Linux/amd64 (gcc) and Windows 
(Visual Studio 10) give different results. Under Linux sheets in the 
repository do not work (in contrast to Windows). The error message is
INFO - ERROR: qlFlatForward: Unable to coerce value from type 
'N5boost10shared_ptrIN13ObjectHandler6ObjectEEE' to type 
'N8QuantLib6HandleINS_5QuoteEEE' - all conversions failed

Help appreciated.

Regards,
  Lars

-- 
Empfehlen Sie GMX DSL Ihren Freunden und Bekannten und wir
belohnen Sie mit bis zu 50,- Euro! https://freundschaftswerbung.gmx.de


-- 
NEU: FreePhone - 0ct/min Handyspartarif mit Geld-zurück-Garantie!		
Jetzt informieren: http://www.gmx.net/de/go/freephone

[Quantlib-dev] coerce problem under Linux/amd64-gcc

[Lars C. <lar...@gm...>]

Hello,

compiling 1.1.0 with calc addins under Linux/amd64 (gcc) and Windows 
(Visual Studio 10) give different results. Under Linux sheets in the 
repository do not work (in contrast to Windows). The error message is
INFO - ERROR: qlFlatForward: Unable to coerce value from type 
'N5boost10shared_ptrIN13ObjectHandler6ObjectEEE' to type 
'N8QuantLib6HandleINS_5QuoteEEE' - all conversions failed

Help appreciated.

Regards,
  Lars

-- 
Empfehlen Sie GMX DSL Ihren Freunden und Bekannten und wir
belohnen Sie mit bis zu 50,- Euro! https://freundschaftswerbung.gmx.de

[Quantlib-dev] Re petitive function calls in discountingbondengine.cpp

[StephenWong <ste...@gm...>]

Within the file ql/pricingengines/bond/discountingbondengine.cpp there is the
DiscountingBondEngine::calculate() method. Within this method, it sets
results_.value and results_.settlementValue by calling the
CashFlows::npv(...) function each time. It does this even if all the
arguments of the two calls are identical. 

Can we not check the arguments first and if they are identical, just call
the CashFlows::npv(...) once? It is a bit of an eye sore for me to see all
the other functions being called within CashFlows::npv(...) over and over
even if there is no such need?

More explicitly, after the line

        results_.value = CashFlows::npv(arguments_.cashflows,
                                        **discountCurve_,
                                        includeRefDateFlows,
                                        results_.valuationDate,
                                        results_.valuationDate);

just insert

if ( includeRefDateFlows == false && results_.valuationDate ==
arguments_.settlementDate )
     results_.settlementValue = results.value;
else {

     results_.settlementValue = CashFlows::npv(arguments_.cashflows,
                                                               
**discountCurve_,
                                                                false,
                                                               
arguments_.settlementDate,
                                                               
arguments_.settlementDate);

}
 
I know this does not look like much but it is a bit of a pain in the neck if
one is going through the code, debugging etc and see all the function calls
being made over and over when there is no need.



-- 
View this message in context: http://old.nabble.com/Repetitive-function-calls-in-discountingbondengine.cpp-tp33026149p33026149.html
Sent from the quantlib-dev mailing list archive at Nabble.com.

Re: [Quantlib-dev] [QuantLib-svn] SF.net SVN: quantlib:[18136] branches/R01020x-branch/QuantLib/ql/patterns/ lazyobject.hpp

[Ferdinando A. <na...@am...>]

On Mon, Dec 19, 2011 at 5:04 PM, Luigi Ballabio
<lui...@gm...> wrote:
> On Mon, Dec 19, 2011 at 4:41 PM,  <na...@us...> wrote:
>> - fixed bug in the update method: non-lazy observers received obsolete data because of calculated_ being still false
>
> I see. I hadn't though of this---probably because I think non-lazy
> observers should be taken out and shot...

I agree with you 100% and I'm trying to shot them all :-D

> Is there any chance you can find the time to write a test case for this?

Yes I can, but not in the coming 2/3 weeks

Re: [Quantlib-dev] enableExtrapolation error?

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

Ok, I've made the change on the 1.2 branch.

Thanks for the heads-up.

Luigi

On Thu, Dec 15, 2011 at 1:27 PM, Luigi Ballabio
<lui...@gm...> wrote:
> On Wed, Nov 23, 2011 at 12:11 AM, barba dos <bar...@ho...> wrote:
>> Hi all, I am looking in the file cashflows.cpp. the piece of code is the
>> ZSpreadFinder constructor.
>>
>> It is about the piece of code that says:
>> curve_(Handle<YieldTermStructure>(discountCurve)
>
>> For the discount curve that is being used, extrapolation is enabled [...] However [...]
>> the extrapolation property is not copied correctly.
>
> Ok, I see.  The above is not all; the complete call is
>
> curve_(Handle<YieldTermStructure>(discountCurve),
>          Handle<Quote>(zSpread_), comp, freq, dc),
>
> curve_ is not a copy of discountCurve, it is a spreaded curve that
> contains discountCurve.
> The extrapolation property of curve_ is never set, so it defaults to
> false.  Try adding
>
> curve_->enableExtrapolation(discountCurve->allowsExtrapolation());
>
> and see if that works for you.
>
> Later,
>    Luigi

Re: [Quantlib-dev] question on discountcurve fitting method

[Bojan N. <bo...@bn...>]

"Han, Guowen" <GH...@dt...> writes:

> Anyone know if there is any  particular reason for using Simplex method in FittedBondDiscountCurve class?


Probably a combination of:

- Simplex does not require derivatives (the Minpack LevenbergMarquardt
  used in QuantLib does not require them either but they are estimated
  internally by finite differences)

- It is slightly more robust to local minima compared to minimisation
  algorithms based on a current and trial point

Best,
Bojan


-- 
Bojan Nikolic          ||          http://www.bnikolic.co.uk

[Quantlib-dev] question on discountcurve fitting method

[Han, G. <GH...@dt...>]

Anyone know if there is any  particular reason for using Simplex method in FittedBondDiscountCurve class?

Thanks,




<BR>_____________________________________________________________
<FONT size=2><BR>
DTCC DISCLAIMER: This email and any files transmitted with it are
confidential and intended solely for the use of the individual or
entity to whom they are addressed. If you have received this email
in error, please notify us immediately and delete the email and any
attachments from your system. The recipient should check this email
and any attachments for the presence of viruses.  The company
accepts no liability for any damage caused by any virus transmitted
by this email.</FONT>

Re: [Quantlib-dev] [QuantLib-svn] SF.net SVN: quantlib:[18136] branches/R01020x-branch/QuantLib/ql/patterns/ lazyobject.hpp

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

On Mon, Dec 19, 2011 at 4:41 PM,  <na...@us...> wrote:
> Revision: 18136
>          http://quantlib.svn.sourceforge.net/quantlib/?rev=18136&view=rev
> Author:   nando
> Date:     2011-12-19 15:41:58 +0000 (Mon, 19 Dec 2011)
> Log Message:
> -----------
> - fixed bug in the update method: non-lazy observers received obsolete data because of calculated_ being still false

I see. I hadn't though of this---probably because I think non-lazy
observers should be taken out and shot...
Is there any chance you can find the time to write a test case for this?

Luigi

Re: [Quantlib-dev] Swedish Calendar: Midsummer Eve

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

Done.  Thanks for the heads-up.

Luigi


On Thu, Dec 15, 2011 at 12:13 PM, Gary Kennedy <gar...@gm...> wrote:
> Yes
>
> Gary
>
> Sent from my iPhone
>
> On 13 Dec 2011, at 10:37, Luigi Ballabio <lui...@gm...> wrote:
>
>> So the limits should be >= 19 and <= 25, right?
>>
>> Luigi
>>
>> On Wed, Dec 7, 2011 at 10:22 PM, Gary Kennedy <gar...@gm...> wrote:
>>> // Midsummer Eve (Friday between June 18-24)
>>>             || (w == Friday && (d >= 18 && d <= 24) && m == June)
>>>
>>> In the Swedish calendar, i think there might be a mistake, is the midsummer
>>> eve not the Friday between June 19-25.
>>>
>>> http://en.wikipedia.org/wiki/Midsummer#Sweden
>>>
>>> I am not an expert on this point, i just noticed it when comparing quantlib
>>> to another system.

Re: [Quantlib-dev] enableExtrapolation error?

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

On Wed, Nov 23, 2011 at 12:11 AM, barba dos <bar...@ho...> wrote:
> Hi all, I am looking in the file cashflows.cpp. the piece of code is the
> ZSpreadFinder constructor.
>
> It is about the piece of code that says:
> curve_(Handle<YieldTermStructure>(discountCurve)

> For the discount curve that is being used, extrapolation is enabled [...] However [...]
> the extrapolation property is not copied correctly.

Ok, I see.  The above is not all; the complete call is

curve_(Handle<YieldTermStructure>(discountCurve),
          Handle<Quote>(zSpread_), comp, freq, dc),

curve_ is not a copy of discountCurve, it is a spreaded curve that
contains discountCurve.
The extrapolation property of curve_ is never set, so it defaults to
false.  Try adding

curve_->enableExtrapolation(discountCurve->allowsExtrapolation());

and see if that works for you.

Later,
    Luigi

Re: [Quantlib-dev] Swedish Calendar: Midsummer Eve

[Gary K. <gar...@gm...>]

Yes 

Gary

Sent from my iPhone

On 13 Dec 2011, at 10:37, Luigi Ballabio <lui...@gm...> wrote:

> So the limits should be >= 19 and <= 25, right?
> 
> Luigi
> 
> On Wed, Dec 7, 2011 at 10:22 PM, Gary Kennedy <gar...@gm...> wrote:
>> // Midsummer Eve (Friday between June 18-24)
>>             || (w == Friday && (d >= 18 && d <= 24) && m == June)
>> 
>> In the Swedish calendar, i think there might be a mistake, is the midsummer
>> eve not the Friday between June 19-25.
>> 
>> http://en.wikipedia.org/wiki/Midsummer#Sweden
>> 
>> I am not an expert on this point, i just noticed it when comparing quantlib
>> to another system.

Re: [Quantlib-dev] Swedish Calendar: Midsummer Eve

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

So the limits should be >= 19 and <= 25, right?

Luigi

On Wed, Dec 7, 2011 at 10:22 PM, Gary Kennedy <gar...@gm...> wrote:
> // Midsummer Eve (Friday between June 18-24)
>             || (w == Friday && (d >= 18 && d <= 24) && m == June)
>
> In the Swedish calendar, i think there might be a mistake, is the midsummer
> eve not the Friday between June 19-25.
>
> http://en.wikipedia.org/wiki/Midsummer#Sweden
>
> I am not an expert on this point, i just noticed it when comparing quantlib
> to another system.

[Quantlib-dev] Swedish Calendar: Midsummer Eve

[Gary K. <gar...@gm...>]

// Midsummer Eve (Friday between June 18-24)
            || (w == Friday && (d >= 18 && d <= 24) && m == June)

In the Swedish calendar, i think there might be a mistake, is the midsummer
eve not the Friday between June 19-25.

http://en.wikipedia.org/wiki/Midsummer#Sweden

I am not an expert on this point, i just noticed it when comparing quantlib
to another system.

gary

Re: [Quantlib-dev] Use code in legacy folder of quantlib

[Klaus S. <kl...@sp...>]

Hi Manas

the code in ql/legacy/libormarketmodels is replaced by the code in 
ql/models/marketmodels. If you start a new project I'd strongly recommend to 
base it on the newer implementation in ql/models/marketmodels.

cheers
 Klaus

On Tuesday 06 December 2011 17:57:41 Luigi Ballabio wrote:
> 2011/12/6 manas bhatt <ma...@ho...>:
> >  I am planning to use the code present in
> > quantlib\ql\legacy\libormarketmodels. Is it ok to use that code or is
> > there new code which can be used in place of the legacy.
>
> The code in ql/models/marketmodels replaced most of it.  I think
> there's still some bits of functionality in the old code that are not
> yet covered by the new, but I don't remember what they are.  Klaus,
> maybe you can chime in?
>
> Luigi
>
> ---------------------------------------------------------------------------
>--- Cloud Services Checklist: Pricing and Packaging Optimization
> This white paper is intended to serve as a reference, checklist and point
> of discussion for anyone considering optimizing the pricing and packaging
> model of a cloud services business. Read Now!
> http://www.accelacomm.com/jaw/sfnl/114/51491232/
> _______________________________________________
> QuantLib-dev mailing list
> Qua...@li...
> https://lists.sourceforge.net/lists/listinfo/quantlib-dev

Re: [Quantlib-dev] Use code in legacy folder of quantlib

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

2011/12/6 manas bhatt <ma...@ho...>:
>  I am planning to use the code present in
> quantlib\ql\legacy\libormarketmodels. Is it ok to use that code or is there
> new code which can be used in place of the legacy.

The code in ql/models/marketmodels replaced most of it.  I think
there's still some bits of functionality in the old code that are not
yet covered by the new, but I don't remember what they are.  Klaus,
maybe you can chime in?

Luigi

[Quantlib-dev] Use code in legacy folder of quantlib

[manas b. <ma...@ho...>]

Hi, I am planning to use the code present in quantlib\ql\legacy\libormarketmodels. Is it ok to use that code or is there new code which can be used in place of the legacy. I am asking this question because there is a test case in testsuite which uses the code present in legacy folder as well a project called MarketModels in quantlib which uses the legacy code.Since the legacy code is suitable for the work i am working on right now, it would be helpful if someone can tell whether it is ok to use the code present in that folderregards,Manas    		 	   		  

Re: [Quantlib-dev] enableExtrapolation error?

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

On Wed, Nov 23, 2011 at 12:11 AM, barba dos <bar...@ho...> wrote:
> Hi all, I am looking in the file cashflows.cpp. the piece of code is the
> ZSpreadFinder constructor.
>
> It is about the piece of code that says:
> curve_(Handle<YieldTermStructure>(discountCurve)
>
> For the discount curve that is being used, extrapolation is enabled. This I
> can confirm by hoovering over discountCurve which tells me:
>
> QuantLib::TermStructure -->>> QuantLib::Extrapolator -->>> { extrapolate_ =
> true }
>
> However, after this has been assigned to the private member curve_ , and I
> hoover over curve_ it shows:
>
> QuantLib::ZeroYieldStructure -->>> QuantLib::YieldTermStructure -->>>
> QuantLib::TermStructure -->>> QuantLib::Extrapolator -->>> { extrapolate_ =
> false }
>
>
> So basically the extrpolation property is not copied correctly.

That's strange.  The curve is not copied at all; assigning to curve_
just copies a pointer, so the two inner TermStructure objects should
be not only identical, but actually the same.  May you try calling the
allowsExtrapolation() method and seeing what it returns, instead of
just hovering over the variable?

Luigi

[Quantlib-dev] [ quantlib-Bugs-3444220 ] ConvertOper string bug

[SourceForge.net <no...@so...>]

Bugs item #3444220, was opened at 2011-11-28 08:25
Message generated for change (Tracker Item Submitted) made by daveblob
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=112740&aid=3444220&group_id=12740

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Dave Blob (daveblob)
Assigned to: Nobody/Anonymous (nobody)
Summary: ConvertOper string bug

Initial Comment:
ConvertOper::operator std::string() const tries to access a temporary variable after it is destroyed (xString becomes a reference to the deleted variable xTemp).  Here's a fix:

    ConvertOper::operator std::string() const {
        const OPER *xString;

        if (oper_->xltype == xltypeStr) {
            xString = oper_;
        } else {
            Xloper xTemp;
            Excel(xlCoerce, &xTemp, 2, oper_, TempInt(xltypeStr));
            xString = &xTemp;
        }
        
	return strConv(xString);
    }

Could be replaced with:

    ConvertOper::operator std::string() const {
        const OPER *xString;

        if (oper_->xltype == xltypeStr) {
            xString = oper_;
	    return strConv(xString);
        } else {
            Xloper xTemp;
            Excel(xlCoerce, &xTemp, 2, oper_, TempInt(xltypeStr));
            xString = &xTemp;
	    return strConv(xString);
        }   
    }

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=112740&aid=3444220&group_id=12740

[Quantlib-dev] [ quantlib-Bugs-3441748 ] Singleton::instance is not thread-safe

[SourceForge.net <no...@so...>]

Bugs item #3441748, was opened at 2011-11-24 02:48
Message generated for change (Tracker Item Submitted) made by mortoray
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=112740&aid=3441748&group_id=12740

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: edA-qa mort-ora-y (mortoray)
Assigned to: Nobody/Anonymous (nobody)
Summary: Singleton::instance is not thread-safe

Initial Comment:
Singleton::instance is not thread-safe in its creation of new Sessions. There is no lock around the access to std::map and thus two threads creating a new session at the same time will conflict. This is a case that actually happens in our code: we manage to get two cores started in their processing at the same time and thus both attempt to create a session at the same time. They do this by having the sessionId return a different key per thread.

There is no way for the end-user to make this function thread-safe as any thread can request the session at any time. This actually makes QuantLib as a whole not thread-safe then, simply because you cannot safely obtain a per-thread session.

The instance() method needs to have a read-write lock so that it works correctly even when creating new sessions.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=112740&aid=3441748&group_id=12740

[Quantlib-dev] enableExtrapolation error?

[barba d. <bar...@ho...>]

Hi all, I am looking in the file cashflows.cpp. the piece of code is the ZSpreadFinder constructor. It is about the piece of code that says: curve_(Handle<YieldTermStructure>(discountCurve) For the discount curve that is being used, extrapolation is enabled. This I can confirm by hoovering over discountCurve which tells me: QuantLib::TermStructure -->>> QuantLib::Extrapolator -->>> { extrapolate_ = true } However, after this has been assigned to the private member curve_ , and I hoover over curve_ it shows: QuantLib::ZeroYieldStructure -->>> QuantLib::YieldTermStructure -->>> QuantLib::TermStructure -->>> QuantLib::Extrapolator -->>> { extrapolate_ = false }  So basically the extrpolation property is not copied correctly. My question: is this a bug or is there a reason for this? If this is on purpose, what would you suggest to turn on extrapolation for the z-spread solver?  Thanks,Barbados.     		 	   		  

Re: [Quantlib-dev] How to copy a Boost::Matrix into a Quantlib::Matrix

[Viala Jean-R. (AMUNDI) <jea...@am...>]

Hi
Thanks I saw it
Your suggestion doesn't work (it doesn't compile) but I think it's
because my mat is not UBLAS 
I will try later but right now the solution below works fine
Regards
JRV 

for (int i=0; i < N; ++i) 
		{
             std::copy(mat[i].begin(), mat[i].end(),
result.row_begin(i));
		}

-----Original Message-----
From: Bojan Nikolic [mailto:bo...@bn...] 
Sent: Wednesday, November 09, 2011 12:39 PM
To: Viala Jean-Renaud (AMUNDI)
Cc: qua...@li...
Subject: Re: [Quantlib-dev] How to copy a Boost::Matrix into a
Quantlib::Matrix


Hi,

"Viala Jean-Renaud (AMUNDI)" <jea...@am...> writes:

> std::copy(mat.begin(), mat.end(), result.begin()); 

Possibly you skimmed over it, but in the snippet I sent this was:

std::copy(mat.begin1(), mat.end1(), result.begin()); 
                   ^ Note the 1 here!
                               ^ here too!

Best,
Bojan

-- 
Bojan Nikolic          ||          http://www.bnikolic.co.uk

Re: [Quantlib-dev] How to copy a Boost::Matrix into a Quantlib::Matrix

[Bojan N. <bo...@bn...>]

Hi,

"Viala Jean-Renaud (AMUNDI)" <jea...@am...> writes:

> std::copy(mat.begin(), mat.end(), result.begin()); 

Possibly you skimmed over it, but in the snippet I sent this was:

std::copy(mat.begin1(), mat.end1(), result.begin()); 
                   ^ Note the 1 here!
                               ^ here too!

Best,
Bojan

-- 
Bojan Nikolic          ||          http://www.bnikolic.co.uk

Re: [Quantlib-dev] How to copy a Boost::Matrix into a Quantlib::Matrix

[Viala Jean-R. (AMUNDI) <jea...@am...>]

Hello

 

It works with the following code:

             

for (int i=0; i < N; ++i) 

{

std::copy(mat[i].begin(), mat[i].end(), result.row_begin(i));

}

With mat being a std::vector<std::vector<double>> and result a Quantlib
Matrix

 

 

We tried 

std::copy(mat.begin(), mat.end(), result.begin()); 

but it did not compile

 

I have a question: do you favors Quantlib::Matrix or
boost::numeric::ublas::matrix<double> for mathematical calculus?

 

Regards

JRV

 

 

      

-----Original Message-----
From: Bojan Nikolic [mailto:bo...@bn...] 
Sent: Tuesday, November 08, 2011 10:02 AM
To: Viala Jean-Renaud (AMUNDI)
Cc: qua...@li...
Subject: Re: [Quantlib-dev] How to copy a Boost::Matrix into a
Quantlib::Matrix

 

 

Hi,

 

jrviala <jea...@am...> writes:

 

> I'm trying to use std::copy but it fails to compile

> It seems a rather simple problem I'm confident someone has used it
somewhere

> could you tell me the correct wording?

 

Something like this should work:

 

 boost::numeric::ublas::matrix<double> m (3, 3);


 QuantLib::Matrix mm(3,3);

 std::copy(m.begin1(), m.end1(), mm.begin());           

 

Although you need to check the column/row convention (I believe this

will transpose the matrix).

 

Best,

Bojan

 

-- 

Bojan Nikolic          ||          http://www.bnikolic.co.uk/ql