On the Relationship among Values of the Same Summary Measure of Error when it is used across Multiple Characteristics at the Same Point in Time: An Examination of MALPE and MAPE
David A. Swanson ()
Additional contact information
David A. Swanson: Department of Sociology, University of California Riverside£¬ Riverside, California 92521, U.S.A.
Review of Economics & Finance, 2015, vol. 5, pages 1-14
This paper deals with an issue that appears to be unexplored by demographers and others who conduct cross-sectional forecasts of populations with multiple characteristics. The issue is based on the question: "if one was conducting an ex post facto evaluation of a forecast or estimate that includes more than one characteristic (e.g., age, gender, race, and geography), how can one explain the fact that there are differences among the summary measures of error for all of the characteristics?" Using a hypothetical demographic forecast as an illustration, the paper examines this issue for two characteristics (race and geography) using a standard summary measure of forecast errors for each of two error dimensions, bias and precision. For the bias dimension, we examine the ¡°Mean Algebraic Percent Error¡± (MALPE) and ¡°Mean Absolute Percent Error (MAPE) for the precision dimension. The paper finds that hitherto unknown relationships across characteristics exist for each of the two summary error measures. For example, in evaluating a forecast that includes more than one characteristic, one can express MAPE taken for one characteristic (e.g., race) in terms of each of the other characteristics (e.g., age and geography). This finding allows one to explain the reason for differences in the summary error measures used to evaluate a forecast or estimate done for more than one characteristic. The findings are informative and suggestive and are likely to be generalizable. While the discussion is not formally rigorous, a formal general proof is presented of the relationship between values of MALPE across multiple characteristics as well as the values of MAPE.
Keywords: Forecasting; Errors; Summary measure reconciliation (search for similar items in EconPapers)
JEL-codes: B4 C8 Z1 (search for similar items in EconPapers)
References: View references in EconPapers View complete reference list from CitEc
Citations Track citations by RSS feed
Downloads: (external link)
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
Persistent link: http://EconPapers.repec.org/RePEc:bap:journl:150301
Ordering information: This journal article can be ordered from
17 Alton Towers Circle, Unit 101 Toronto, ON, M1V3L8, Canada
Access Statistics for this article
Review of Economics & Finance is currently edited by H. Carlson
More articles in Review of Economics & Finance from Better Advances Press, Canada 17 Alton Towers Circle, Unit 101 Toronto, ON, M1V3L8, Canada.
Series data maintained by Carlson ().