1: \begin{abstract}

2: We propose a goodness-of-fit test for the distribution of errors from

3: a multivariate indirect regression model. The test statistic is based

4: on the Khmaladze transformation of the empirical process of

5: standardized residuals. This goodness-of-fit test is consistent at the

6: root-$n$ rate of convergence, and the test can maintain power against

7: local alternatives converging to the null at a root-$n$ rate.

8: \end{abstract}

9: