We describe different methods for estimating the bispectrum of Cosmic Microwave Background data .
In particular we construct a minimum variance estimator for the flat-sky limit and compare results with previously-studied frequentist methods .
Application to the MAXIMA dataset shows consistency with primordial Gaussianity .
Weak quadratic non-Gaussianity is characterised by a tunable parameter f _ { NL } , corresponding to non-Gaussianity at a level \sim 10 ^ { -5 } f _ { NL } ( ratio of non-Gaussian to Gaussian terms ) , and we find limits of |f _ { NL } | < 950 for the minimum-variance estimator and |f _ { NL } | < 1650 for the usual frequentist estimator .
These are the tightest limits on primordial non-Gaussianity which include the full effects of the radiation transfer function .