Sensitivity bounds for use with flawed data Original Research Article

Data do not adhere to the usual statistical assumptions and applied work is made difficult by the need to cope with the myriad of problems that arise from flawed data. Prominent examples are data that contain missing values or where variables are only available in categorised form. Standard solutions include omitting incomplete records and replacing missing or categorised observations by some representative values. In these and other cases of flawed data, there is some information available on the potential range into which any particular observation could feasibly fall. We contend that there is considerable diagnostic value in exploiting this information to compute bounds for the coefficient estimates and related statistics such as t-ratios. While one of the standard solutions may ultimately be used to produce a set of estimates, the bounds provide an indication of how sensitive these results are to the particular solution chosen. This approach is developed and illustrated by way of several examples.