We discuss methods for calculating multivariate normal probabilities by simulation and two new Stata programs for this purpose: mdraws for deriving draws from the standard uniform density using either Halton or pseudorandom sequences, and an egen function, mvnp(), for calculating the probabilities themselves. Several illustrations show how the programs may be used for maximum simulated likelihood estimation.
HaanP., and UhlendorffA.2006. Estimation of multinomial logit models with unobserved heterogeneity using maximum simulated likelihood. Stata Journal6: 229–245.
9.
HessS., and PolakJ.2003. An alternative method to the scrambled Halton sequence for removing correlation between standard Halton sequences in higher dimensions. Paper presented at the 2003 European Regional Science Conference, Jyväskylä, Finland. http://www.jyu.fi/ersa2003/cdrom/papers/406.pdf.
JenkinsS. P., CappellariL., LynnP., JäckleA., and SalaE.Forthcoming. Patterns of consent: evidence from a general household survey. Journal of the Royal Statistical Society, Series A169. http://www.iser.essex.ac.uk/pubs/workpaps/pdf/2004-27.pdf.
12.
SándorZ., and AndrásP.2004. Alternative sampling methods for estimating multivariate normal probabilities. Journal of Econometrics120: 207–234.
13.
SternS.1997. Simulation-based estimation. Journal of Economic Literature35: 20062039.
14.
StewartM.2006. Maximum simulated likelihood estimation of random-effects dynamics probit models with autocorrelated errors. Stata Journal6: 256–272.
15.
TrainK.2003. Discrete Choice Methods with Simulation.Cambridge: Cambridge University Press.
