While endpoint impedance of manipulators can be modulated through feedback control, it may not be feasible to reduce it below a certain bound that is dictated by the inherent mechanical characteristics of the robot. In applications where the endpoint impedance of a robot arm needs to be modulated within a large range, it is beneficial to have a method that can estimate the passive characteristics (endpoint inertia and endpoint damping) of the robot arm as early as in the design phase, and use this information to modify the design, if necessary, before building the prototype. In our previous work, we developed a numerical simulation method to estimate 1-D endpoint impedance that is applicable only to limited scenarios. In this article, we present a new numerical simulation method to estimate the 2-D endpoint impedance for any general 2-D manipulators at various locations inside the workspace. We show that our method accurately estimates the endpoint inertia and presents feasible estimates of endpoint damping through numerical simulations. An analysis of the feasibility of using the robot for interaction tasks - based on the obtained impedance characteristics - is also discussed. With our method, robot output impedance can be tailored as early as in the design phase.