Dynamic adaptation of transmission power has been researched as a technique for improving the performance and capacity of wireless networks. In this paper an estimator-based algorithm is presented for distributed power control. The proposed power control policy is optimal with respect to users dynamically allocating transmit power so as to minimize an objective function consisting of the user's performance degradation and the network interference. The policy enables a user to address various user-centric and network-centric objectives by adapting power in either a greedy or energy efficient manner. The algorithm is predictive, with a user performing autonomous interference estimation and prediction prior to adapting transmit power. Also, closed-loop implementation of the algorithm is of reasonable complexity thus allowing for distributed online operation. Subsequently, the robustness of the algorithm to stochastic detriments such as a time varying channel and noisy measurements is investigated.