1: \begin{abstract}

2: A class of interior point methods using inexact directions is analysed. The

3: linear system arising in interior point methods for linear programming is

4: reformulated such that the solution is less sensitive to perturbations in the

5: right-hand side. For the new system an implementable condition is formulated

6: that controls the relative error in the solution. Based on this condition, a

7: feasible and an infeasible potential reduction method are described which retain

8: the convergence and complexity bounds known for exact directions.

9: %\keywords{Interior point methods \and Inexact directions \and Error control}

10: \end{abstract}