An Asymptotically Distribution-Free Test for Symmetry versus Asymmetry

Randles, Ronald H.; Fligner, Michael A.; Policello, George E.; Wolfe, Douglas A.

doi:10.1080/01621459.1980.10477448

Cited by 149 publications

(41 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Therefore, a non-parametric triples test, first developed by Randles et al (1980), and introduced in the economics literature by Verbrugge (1997), has been suggested as an alternative, more powerful test to derive cyclical asymmetry (Razzak, 2001;Verbrugge, 1997).…”

Section: Appendix B: Non Parametric Triples Testmentioning

confidence: 99%

See 1 more Smart Citation

Weathering Tight Economic Times: The Sales Evolution of Consumer Durables Over the Business Cycle

et al. 2004

View full text Add to dashboard Cite

Despite their obvious importance, not much marketing research focuses on how business-cycle fluctuations affect individual companies and/or industries. Often, one only has aggregate information on the state of the national economy, even though cyclical contractions and expansions need not have an equal impact on every industry, nor on all firms in that industry. Using recent time-series developments, we introduce various measures to quantify the extent and nature of business-cycle fluctuations in sales. Specifically, we discuss the concept of cyclical volatility, and derive a dynamic comovement elasticity between the economy as a whole and the cyclical fluctuations in various performance series. To further enhance our understanding of how consumers adjust their purchasing behavior across different phases of the business cycle, two other notable features related, respectively, to the relative size of the peaks and troughs and the rate of change in upward and downward parts of the cycle, are explicitly considered. Of specific interest in this respect are the notion of deepness and steepness asymmetry. We apply these concepts to a broad set (24) of consumer durables, for which we analyze the cyclical sensitivity in their sales evolution. In that way, we (i) derive a novel set of empirical generalizations, and (ii) test different marketing theory-based hypotheses on the underlying drivers of cyclical sensitivity. Consumer durables are found to be more sensitive to business-cycle fluctuations than the general economic activity, as expressed in an average cyclical volatility of more than four times the one in GNP, and an average dynamic comovement elasticity in excess of 2. This observation calls for an explicit consideration of cyclical variation in durable sales. Interestingly, the combined evidence across all durables suggests that asymmetry is present in the speed of up- and downward movements, as durable sales fall much quicker during contractions than they recover during economic expansions. Finally, key variables related to the industry’s pricing activities, the nature of the durable (convenience vs. leisure), and the stage in a product’s life cycle tend to moderate the extent of cyclical sensitivity in durable sales patterns. Copyright Springer Science+Business Media, Inc. 2004business cycles, sales evolution, consumer durables, time-series econometrics,

show abstract

Section: Appendix B: Non Parametric Triples Testmentioning

confidence: 99%

“…We refer to the study of Randles et al (1980) for both a more elaborate discussion/description of the methodology, and for the derivation ofσ 2 η .…”

Section: Appendix B: Non Parametric Triples Testmentioning

confidence: 99%

Weathering Tight Economic Times: The Sales Evolution of Consumer Durables Over the Business Cycle

et al. 2004

View full text Add to dashboard Cite

show abstract

“…These tests are the entropybased test of Maasoumi (2007, 2008), the Triples test developed by Randles et al (1980), and the conditional symmetry test suggested by Bai and Ng (2001).…”

Section: Methodsmentioning

confidence: 99%

“…To achieve this goal, we concentrate on the G7 countries and consider both residential and industrial electricity demand. Our econometric approach is varied in the sense that we use three different methods: the new entropy-based test for conditional symmetry proposed by Maasoumi (2007, 2008), the Triples test developed by Randles et al (1980), and the conditional symmetry test suggested by Bai and Ng (2001). We found strong evidence that both residential and industrial electricity demand have symmetric distributions.…”

Section: Introductionmentioning

confidence: 90%