QuantLib › quantlib-users

Confusing behavior from BinomialVanillaEngine constructor

Classic

List

Threaded

2 messages Options

alex

Confusing behavior from BinomialVanillaEngine constructor

Hi,

For some reason, this line [0] is crashing with a length_error when reserving space for a vector<Time>. I originally thought the error may be originating in the NPV calculation, but the protected timeSteps_ variable in BinomialVanillaEngine is actually not being set to the expected value.

This method [1] is where the number of time steps is set. Passing in a value of 1025 (or 0 -- the value didn't seem to matter) results in the BinomialVanillaEngine constructor setting timeSteps_ to a number that's about 1.475e19. (This is on a 64-bit machine. Similar behavior occurred when I did a 32-bit build, but I didn't note the values.)

Is this a known bug, or is something going wrong somewhere on my end?

[0] https://gist.github.com/aml3/65038735906fd2b14e1f#file-pricing-cpp-L54

[1] https://gist.github.com/aml3/65038735906fd2b14e1f#file-pricing-cpp-L3

Thanks,

Alex Lamana

------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users

alex

Re: Confusing behavior from BinomialVanillaEngine constructor

Bug was on my end. Forgot to initialize a value.

Sorry for the spam.

On Wed, Mar 18, 2015 at 11:11 AM, Alexander Lamana <[hidden email]> wrote:

Hi,

For some reason, this line [0] is crashing with a length_error when reserving space for a vector<Time>. I originally thought the error may be originating in the NPV calculation, but the protected timeSteps_ variable in BinomialVanillaEngine is actually not being set to the expected value.

This method [1] is where the number of time steps is set. Passing in a value of 1025 (or 0 -- the value didn't seem to matter) results in the BinomialVanillaEngine constructor setting timeSteps_ to a number that's about 1.475e19. (This is on a 64-bit machine. Similar behavior occurred when I did a 32-bit build, but I didn't note the values.)

Is this a known bug, or is something going wrong somewhere on my end?

[0] https://gist.github.com/aml3/65038735906fd2b14e1f#file-pricing-cpp-L54
[1] https://gist.github.com/aml3/65038735906fd2b14e1f#file-pricing-cpp-L3

Thanks,
Alex Lamana