MAPE-R: an empirical comparison with MAPE

Thomas M. Bryan, Third Wave Research

MAPE is simple to calculate and easy to understand, which attest to its popularity, but it does not meet all of the criteria needed for a summary measure of error. To overcome the deficiencies in MAPE, MAPE-R was developed, but it has yet been extensively compared with MAPE empirically. This paper makes an extensive empirical comparison between MAPE and MAPE-R using block group level population estimates from a private Application Service Provider (ASP) with demographic expertise in the United States. The test involves comparing the private vendor’s 2000 block group population estimates with a large sample of the 208,668 Census 2000 block groups in the United States. The exercise will demonstrate the utility of the MAPE-R in comparison with MAPE as a high-quality summary measure of error in small-area population estimates.