Some Continuous Univariate Right-skewed Probability Distributions and Their Applications on Real Data

Authors

Ahmed A. NgarmboFollow

Document Type

Thesis

Degree

Master of Science (MS)

Major/Program

Statistics

First Advisor's Name

B.M.Golam Kibria

First Advisor's Committee Title

Committee Chair

Second Advisor's Name

Florence George

Second Advisor's Committee Title

Co-Committee Chair

Third Advisor's Name

Sneh Gulati

Third Advisor's Committee Title

Committee member

Keywords

Right-skewed probability, Rmax, IMR

Date of Defense

6-16-2023

Abstract

Probability theory includes various continuous probability distributions. Depending on their properties and characteristics, some distributions fit data better. The process of identifying a suitable distribution involves the assessment of multiple distributions. This research reviews some continuous right-skewed probability distributions to real-world data including U.S. infant mortality from 1950-2023, Florida's unemployment rate from 1992 2021, and the radius of maximum wind from 1908-2021. Twelve continuous probability distributions were evaluated: Normal, Exponential, Gamma, Log-normal, Gompertz, Gumbel, Chi-squared, Generalized extreme value, Weibull, Pareto, Rayleigh, and Lomax. The maximum likelihood method (MLE) was used to estimate the distribution's parameters. In addition, a graphical technique was utilized to visually assess a fitted distribution. The results showed that the Generalized Extreme Value distribution and the Gamma distribution are the best-fitted probability distributions for Florida's unemployment rate and the radius of maximum wind, respectively, while the Pareto distribution is the best-fitted probability distribution for the U.S. infant mortality rate.

Identifier

FIDC011151

Recommended Citation

Ngarmbo, Ahmed A., "Some Continuous Univariate Right-skewed Probability Distributions and Their Applications on Real Data" (2023). FIU Electronic Theses and Dissertations. 5424.
https://digitalcommons.fiu.edu/etd/5424