A Comparison of Modality Tests Based on Real Life Data Applications: A Comparison of Modality Tests Based on Real Life Data Applications

Rana Muhammad Imran  Arshad; Shahina  Imran; Farrukh  Jamal; Gamze  Ozel

Authors

Rana Muhammad Imran Arshad Department of Statistics Govt., S.E. College, Bahawalpur, Pakistan
Shahina Imran Department of Economics Govt., S. E. College, Bahawalpur, Pakistan
Farrukh Jamal Department of Statistics Govt., S.A. Postgraduate College, Dera Nawab Sahib, Bahawalpur, Pakistan
Gamze Ozel Department of Statistics, Hacettepe University, Ankara, Turkey

Keywords:

Nonparametric tests, Silverman’s bandwidth test, Hartigan’s dip test, Proportional Mass (PM) test, Excess Mass test, unimodality, Real life data application

Abstract

Nonparametric tests for the modality are known as distribution-free tests to evaluate the evidence about homogeneity in a population. Silverman’s bandwidth test by Silverman [1], the Hartigan’s dip test by Hartigan [6],Hartigan and Hartigan [7], Proportional Mass (PM) test by Cavallo and Ringobon [4] and Excess Mass (EM) test by Muller and Sawitzki [9] are very popular modality tests in literature. The researchers of these tests claimed that their tests perform well to detection of unimodality, bimodality and multimodality. This study focusses on the comparison of these tests based on real data applications. The results show that Silverman’s bandwidth test is the best test for detecting modality whether it is unimodality, bimodality or multimodality.

References

Aslam, S. Impact of Microfinance On Poverty Alliviation: A Case Study of Bahawalpur District, Allama Iqbal Open University Islamabad, PhD thesis (1999).

Bianchi, M. Testing for convergence: evidence from nonparametric multimodality tests, Journal of Applied Econometrics,12(4): 393-409 (1997).

Cheng, M, & P. Hall, Calibrating the excess mass and dip tests of modality, Journal of the Royal Statistical Society, Series B, 60: 579-590(1998).

Cavallo, A, & R. Rigobon, The Distribution of the size of price changes, NBER Working Paper No. 16760, 1-40(2014).

Hall, P, & M. York, On the calibration of Silverman’s test for multimodality. Statistica Sinica, 11: 515– 536(2001).

Hartigan, P.M, Algorithm AS217: Computation of the Hartigan Dip Test Statistic to test for unimodality, Appl. Statist, 34: 320–325(1985).

Hartigan, P.M., & J.A Hartigan, the Hartigan Dip Test of unimodality, Ann Statist, 13(1): 70-84(1985).

Hall, P, & S. OOI, S, Attributing a Probability to the Shape of a Probability Density, The Annals of Statistics, Institute of Mathematical Statistics, 32(5): (2004).

Mueller, D.W. & G. Sawitzki, Excess mass estimates and tests for multimodality, JASA, 86: 738 -746(1991).

Silverman, B.W. Some properties of a test for multimodality based on kernel density estimates, Probability, Statistics, and Analysis, Cambridge University Press, Cambridge (1983).

Silverman, B.W, Density Estimation for Statistics and Data Analysis, Monographs on Statistics and Applied Probability 26, Chapman and Hall, London (1986).

Ling, X., E. Bedrick., T. Hanson., C. Restrepo, A Comparison of Statistical Tools for Identifying Modality in Body Mass Distributions, Journal of Data Science, 12:175-196(2014).

A Comparison of Modality Tests Based on Real Life Data Applications