Ensuring data quality is essential for producing accurate and reliable federal statistics. A common approach is to use ratio edits, which are widely used in economic and establishment surveys in order to identify records with implausible relationships between variables. A ratio edit checks if ratios are within specified limits, which can be either symmetric or asymmetric depending on the data distribution. In a multivariate setting when a component ratio is flagged as unusual, it does not mean the entire data vector is incorrect; it suggests that some of the components may be inconsistent. Traditional methods in the multivariate case often use the Mahalanobis distance; however, this approach relies on the multivariate normality assumption and the region derived, being ellipsoidal, does not provide interpretable bounds for each variable. To address these limitations, this work presents a framework for multivariate ratio edits based on tolerance regions (TRs) of rectangular shape under both parametric and nonparametric scenarios. The regions so derived will include a specified proportion or more of the population with a given confidence level, and the rectangular shape provides separate bounds for each variable. The parametric setup considered is based on the multivariate normality assumption, and trimming is used to reduce the influence of outliers in the ratio data. In the nonparametric setting, data depths and Statistically Equivalent Blocks (SEBs) are used to obtain distribution-free TRs. Results from Monte Carlo simulations are reported in order to evaluate Type I/II error rates and TR volumes under different contaminations of the data, differing dimensions, and different parameter choices. Results indicate that rectangular central TRs excel across different scenarios, whereas simultaneous central TI is a default choice when a mixture of two sided and one sided outlier flagging is desired. In the nonparametric setup, SEB appears to be preferable. The methodologies are demonstrated using the Annual Survey of Manufactures (ASM) microdata for 2018–2021, U.S. Census Bureau [2025].