Posted by
Lapin on
Sep 25, 2009; 2:21pm
URL: http://quantlib.414.s1.nabble.com/QuantLibAddin-and-Monte-Carlo-simulation-tp8185.html
Hi all,
I am trying to expose MCBarrierEngine into quantLibXL but I am really struggling to pass the PseudoRandomNumber generator...
Let appart the relevant includes, in my view you would need to add the following code in pricingengine.cpp
template<typename T>
class MCBarrierEngine : public PricingEngine {
public:
MCBarrierEngine(
const boost::shared_ptr<ObjectHandler::ValueObject>& properties ,
const boost::shared_ptr<QuantLib::GeneralizedBlackScholesProcess>& process ,
const QuantLib::Size maxTimeStepsPerYear ,
bool brownianBridge ,
bool antitheticVariate ,
bool controlVariate ,
const QuantLib::Size requiredSamples ,
const QuantLib::Real requiredTolerance ,
const QuantLib::Size maxSamples ,
bool isBiased ,
const T& RSQ ,
const QuantLib::BigNatural seed ,
bool permanent );
And this in pricingengine.cpp
template<typename T>
MCBarrierEngine<T>::MCBarrierEngine(
const boost::shared_ptr<ObjectHandler::ValueObject>& properties ,
const boost::shared_ptr<QuantLib::GeneralizedBlackScholesProcess>& process ,
const QuantLib::Size maxTimeStepsPerYear ,
const bool brownianBridge ,
const bool antitheticVariate ,
const bool controlVariate ,
const QuantLib::Size requiredSamples ,
const QuantLib::Real requiredTolerance ,
const QuantLib::Size maxSamples ,
const bool isBiased ,
const T& RSQ ,
const QuantLib::BigNatural seed ,
bool permanent )
: PricingEngine(properties, permanent)
{
libraryObject_ = boost::shared_ptr<QuantLib::PricingEngine>(new
QuantLib::MCBarrierEngine<RSQ>(process,
maxTimeStepsPerYear, brownianBridge,
antitheticVariate, controlVariate,
requiredSamples, requiredTolerance,
maxSamples, isBiased, seed)) ;
}
To finish, you need to add the following constructor in pricingengine.xml
<Constructor name='qlMCBarrierEnginer'>
<libraryFunction>MCBarrierEngine</libraryFunction>
<SupportedPlatforms>
<SupportedPlatform name='Excel'/>
</SupportedPlatforms>
<ParameterList>
<Parameters>
<Parameter name='ProcessID' >
<type>QuantLib::GeneralizedBlackScholesProcess</type>
<superType>libraryClass</superType>
<tensorRank>scalar</tensorRank>
<description>GeneralizedBlackScholesProcess object ID.</description>
</Parameter>
<Parameter name='TimeStepPerYear'>
<type>long</type>
<tensorRank>scalar</tensorRank>
<description>Time Steps Per Year</description>
</Parameter>
<Parameter name='BrownianBridge' default='FALSE'>
<type>bool</type>
<tensorRank>scalar</tensorRank>
<description>Using a Brownian Bridge?</description>
</Parameter>
<Parameter name='AntitheticVariate' default='FALSE'>
<type>bool</type>
<tensorRank>scalar</tensorRank>
<description>Using a Antithetic Variate?</description>
</Parameter>
<Parameter name='ControlVariate' default='FALSE'>
<type>bool</type>
<tensorRank>scalar</tensorRank>
<description>Using a Control Variate?</description>
</Parameter>
<Parameter name='RequiredSamples'>
<type>long</type>
<tensorRank>scalar</tensorRank>
<description>Minimum number of pathes</description>
</Parameter>
<Parameter name='RequiredTolerance'>
<type>long</type>
<tensorRank>scalar</tensorRank>
<description>Tolerance on the error</description>
</Parameter>
<Parameter name='MaxSamples'>
<type>long</type>
<tensorRank>scalar</tensorRank>
<description>Maximum number of pathes</description>
</Parameter>
<Parameter name='RandomNumberGenerator'>
<type>QuantLib::PseudoRandomSequenceGenerator</type>
<superType>underlyingClass</superType>
<tensorRank>scalar</tensorRank>
<description>Random Number Generator</description>
</Parameter>
<Parameter name='Seed' default='123'>
<type>long</type>
<tensorRank>scalar</tensorRank>
<description>Seed used for the random numbers</description>
</Parameter>
</Parameters>
</ParameterList>
</Constructor>
I get an error C3203: 'PseudoRandomSequenceGenerator' : unspecialized class template can't be used as a template argument for template parameter 'T', expected a real type ... since I use a non unspecialized template class for the generator.
Is there a way to do 1 XL function for all the Pseudo Random Generators?
Cheers