Simultaneous consideration of both demand and price uncertainties has not been studied extensively in the literature. This problem is mathematically intractable for cases where a complex problem structure exists. This paper addresses the multi-period single-item lot sizing problem with stochastic demand and price on a rolling horizon basis. Problem formulation permits lost sale and backordering. Unit holding cost depends on purchasing price. In this study, we propose three new lot sizing heuristics based on a rolling horizon for this problem. The first two heuristics are the modified versions of well-known silver-meal and least-unit cost heuristics. The last heuristic known as cost-benefit (CB) is based on a cost-benefit evaluation at decision points. An extensive simulation analysis is performed for different values of set-up cost, number of rolling horizon periods, coefficient of variation of demand and price, backorder ratio and service level factors. Simulation study also considers different demand and price scenarios. The proposed heuristics are compared with each other by considering different cost components and they are also compared with deviation from the Wagner-Whitin solution. Simulation experiments show that the proposed CB heuristic outperforms the other two heuristics in most of the scenarios.