A New One Parameter Unit Distribution: Median Based Unit Rayleigh (MBUR): Parametric Quantile Regression Model

Iman Mohamed Attia

2024-10-19

Abstract:Parametric quantile regression is illustrated for the one parameter new unit Rayleigh distribution called Median Based Unit Rayleigh distribution (MBUR) distribution. The estimation process using re-parameterized maximum likelihood function is highlighted with real dataset example. The inference and goodness of fit is also explored.

Methodology,Probability

What problem does this paper attempt to address?

### What problem does this paper attempt to solve?

This paper aims to solve the problem of exploring the relationship between covariates and response variables in highly skewed - distributed data. Specifically, the author proposes a new one - parameter Median Based Unit Rayleigh (MBUR) distribution and uses this distribution for parametric quantile regression analysis.

#### Main problems and solutions:
1. **The influence of data skewness and outliers**:
- In many practical applications, the distribution of the response variable may be highly skewed and there may be outliers. These problems will cause the assumptions in the least - squares regression model (such as normality and homoscedasticity) to be violated, thus affecting the estimation and inference process of the regression coefficients.
- Parametric quantile regression is a robust alternative method that can handle cases where these assumptions are violated. It models through conditional quantiles rather than conditional means, so it is more robust to skewness and outliers.

2. **Limitations of existing models**:
- Many existing regression models (such as Beta regression, Kumaraswamy regression, etc.) are suitable for data within the unit interval, but they usually do not have a closed - form quantile function, which limits their application in quantile regression.
- The MBUR distribution has a closed - form quantile function, so it can be used for parametric quantile regression, thereby better describing the relationship between covariates and response variables.

3. **Application of the new distribution**:
- The author proposes the MBUR distribution and shows how to use this distribution for quantile regression analysis. By re - parameterizing the probability density function (PDF) and the cumulative distribution function (CDF), the author applies the MBUR to actual data sets and verifies its effectiveness in handling skewed data and outliers.

4. **Model evaluation and selection**:
- To evaluate the goodness - of - fit of the MBUR model, the author uses Randomized Quantile Residuals and Cox - Snell residuals and compares different model selection criteria (such as AIC, BIC, etc.). The results show that the MBUR model is superior to other models in many aspects, especially when expressing the relationship between explanatory variables and response variables.

#### Summary:
This paper provides a new parametric quantile regression method by introducing the MBUR distribution, which solves the problem of violating the assumptions of traditional regression models in highly skewed and outlier - existing data. The closed - form quantile function of the MBUR distribution makes it an effective tool that can more accurately describe the relationship between covariates and response variables.