wilmott.com

Most of the models for interest rates ( for instance I'm using a G2++ model) admits with a positive probability negatives rates. Without chosing ontother model, how can we deal wioth this problem?Taking, for example, the absolute value of the genarted paths will not cause the loss of no-arbitrage in the model? The usage of such a kind of model will not cause problems in the pricing of derivative securities trhough Monte Carlo simualtions?Regards

I am slightly out of topic. However, I think negative IR make economic sense. In the past, negative IR have been experienced - for example, if I remember correctly, in Switzerland during WW2: interests were charged by the Federation to hold money in safe chocolate vaults.