The identification of reduced-form VAR models has been the subject of numerous debates in the literature. Different sets of identifying assumptions can lead to very different conclusions regarding the effects of shocks. This paper proposes a theoretically consistent identification strategy using restrictions implied by a DSGE model. Monte Carlo simulations suggest that both quantitative and qualitative restrictions work well together, where they act as complements to each other, in minimizing errors in finding the correct VAR identification. When using misspecified model restrictions, the data tend to push the identified VAR responses away from the misspecified model and closer to the true data-generating process.