The Odd Log-Logistic Student t Distribution: Theory and Applications
- 229 Downloads
The normal distribution is most used in analysis of experiments. However, it is not suitable to apply in situations where the data have evidence of bimodality or heavier tails than the normal distribution. So, we propose a new four-parameter model called the odd log-logistic Student t distribution as an alternative to the normal and Student t distributions. The new distribution can be symmetric, platykurtic, mesokurtic or leptokurtic and may be unimodal or bimodal. Its various structural properties can be determined from the linear representation of its density function. The estimation of the model parameters is performed by maximum likelihood. The proposed distribution can be used as an alternative for randomized complete block design, thus providing analysis of real data more realistic than other special regression models. We perform a sensitivity analysis to detect influential or outlying observations, and construct generated envelopes from the residuals to select appropriate models. We illustrate the importance of the proposed model by means of three real data sets in analysis of experiments carried out in different regions of Brazil.
KeywordsLog-logistic distribution Maximum likelihood estimation Moment Regression model Student t distribution
We are very grateful to a referee and an associate editor for helpful comments that considerably improved the paper. We gratefully acknowledge financial support from CAPES, CNPq and FAPAC.
- Doornik, J.A. (2006). An Object-Oriented Matrix Language - Ox 4. 5th ed. Timberlake Consultants Press, London.Google Scholar
- Dunn. P.K., Smyth G.K. (1996). Randomized quantile residuals. Journal Computational and Graphical Statistics, 5, 236-244.Google Scholar
- Lima, M.C.S, Cordeiro, G.M., Ortega, E.M.M. (2015). A new extension of the normal distribution. Journal of data Science, 13, 385-408.Google Scholar
- Shirke, D.T., Kakade, C. S. (2006). On exponentiated log-normal distribution. International Journal of Agricultural and Statistical Science, 2, 319-326.Google Scholar