1: \begin{abstract}

2: An efficient conditioning technique, the so-called Brownian Bridge

3: simulation, has previously been applied to eliminate pricing bias that

4: arises in applications of the standard discrete-time Monte Carlo method to

5: evaluate options written on the continuous-time extrema of an underlying

6: asset. It is based on the simple and easy to implement analytic formulas for

7: the distribution of one-dimensional Brownian Bridge extremes. This paper

8: extends the technique to the valuation of multi-asset options with knock-out

9: barriers imposed for all or some of the underlying assets. We derive formula

10: for the unbiased option price estimator based on the joint distribution of

11: the multi-dimensional Brownian Bridge dependent extrema. As analytic

12: formulas are not available for the joint distribution in general, we develop

13: upper and lower biased option price estimators based on the distribution of

14: independent extrema and the Fr\'{e}chet lower and upper bounds for the

15: unknown distribution. All estimators are simple and easy to implement. They

16: can always be used to bind the true value by a confidence interval.

17: Numerical tests indicate that our biased estimators converge rapidly to the

18: true option value as the number of time steps for the asset path simulation

19: increases in comparison to the estimator based on the standard discrete-time

20: method. The convergence rate depends on the correlation and barrier

21: structures of the underlying assets.

22: \end{abstract}

23: