In this paper, we propose new mathematical programming approaches for computing time-dependent bid prices in network revenue management problems. In contrast with previous models, ours can accommodate more than one customer request between two successive bid price updates, as frequently occurs in practice. As a first approach, we introduce a simplified version of our time-dependent bid price model in which the random demand is replaced by its expectation in each time period. Next, we propose three heuristic scenario-based stochastic programming methods, whose aim is to improve robustness to uncertainty in the case of stochastic demand. The time-dependent bid prices are found by solving mixed-integer linear programs. In our numerical experiments, our proposed methods outperform the previously-proposed techniques by 2 to 3% on average, in terms of average revenue under the dynamic strategies computed by the methods.