I am calibratting Gaussian one factor model (reduced from g2++ by making second constant zero) using ATM Eur Cap volatilities.downloaded eur swap rates .using VCUB command on bloomberg to get cap volatilities . is this right source of data ?.after calibrating did the monte carlo simulationmonte carlo simulation of EUR 3m is giving me weird values ? because my value of "a" is coming very small . Any idea of what should be range of "a" in G2++ model

QuoteOriginally posted by: CloudNineI am calibratting Gaussian one factor model (reduced from g2++ by making second constant zero) using ATM Eur Cap volatilities.downloaded eur swap rates .using VCUB command on bloomberg to get cap volatilities . is this right source of data ?.after calibrating did the monte carlo simulationmonte carlo simulation of EUR 3m is giving me weird values ? because my value of "a" is coming very small . Any idea of what should be range of "a" in G2++ modelNot sure how you are fitting it, but if you have set the mean reversion factor for one of the processes to 0, then in effect the stochastic mean is just 'wandering'. In that case, I can imaging that in order to match the variance of the process the mean reversion factor fo the second process is squashed to zero.If you are not interested in one of the factors, could you just use and OU processes with chosen mean and fitted to the variance?