Simultaneous parameter estimation and variable selection via the logit-normal continuous analogue of the
W Thomson1, S Jabbari1,2, A E Taylor3,4
11 School of Mathematics, University of Birmingham , Birmingham , UK.
Journal of the Royal Society, Interface
|April 9, 2019
Summary
We developed a new Bayesian prior, the logit-normal distribution, for flexible statistical modeling. This method enhances parameter estimation and variable selection in biological data analysis, matching machine learning performance.
Area of Science:
- Statistics
- Bioinformatics
- Machine Learning
Background:
- Bayesian inference is crucial for statistical modeling.
- Spike-and-slab priors are common but have limitations.
- Flexible priors are needed for complex data.
Purpose of the Study:
- Introduce a novel Bayesian prior: the logit-normal distribution.
- Demonstrate its utility in parameter estimation and variable selection.
- Compare its performance against existing methods.
Main Methods:
- Developed the logit-normal prior as a continuous analogue of spike-and-slab.
- Applied the prior in a simulation study.
- Validated the prior using metabolomics and genomics datasets.
Main Results:
- The logit-normal prior facilitates flexible parameter estimation.
- It enables effective variable and model selection.
- Performance is comparable to machine learning methods in generalization.
Conclusions:
- The logit-normal prior offers a powerful tool for statistical analysis.
- It integrates well with classical, interpretable models.
- Provides a robust alternative for biological data analysis.
Related Concept Videos
01:26Distributions to Estimate Population Parameter
5.1K
The accurate values of population parameters such as population proportion, population mean, and population standard deviation (or variance) are usually unknown. These are fixed values that can only be estimated from the data collected from the samples. The estimates of each of these parameters are sample proportion, the sample mean, and sample standard deviation (or variance). To obtain the values of these sample statistics, data are required that have particular distribution and central...
5.1K
02:32Variation: Normal Distribution, Range, and Standard Deviation
27.0K
In the field of psychology, there are several ways to organize measurements of a trait, feature, or characteristic (i.e., variables). Qualitative data, such as ethnicity, can be tabulated into a frequency count to provide information about the proportion, as well as the variety of groups in a sample or population. On the other hand, researchers can perform a wider set of calculations on quantitative data. The mean, mode, and median, for instance, are central tendency measures to identify a...
27.0K
01:11Variability: Analysis
471
Measures of variability are statistical metrics that reveal the dispersion pattern within a dataset. They are pivotal in biostatistics, providing insights into the heterogeneity within health and biological data. Variability signifies the degree to which data points diverge from one another, helping researchers understand the potential range of values and associated uncertainty within the data.
The range is a simple measure of variability, indicating the difference between the highest and...
The range is a simple measure of variability, indicating the difference between the highest and...
471
01:09Random Variables
17.6K
A random variable is a single numerical value that indicates the outcome of a procedure. The concept of random variables is fundamental to the probability theory and was introduced by a Russian mathematician, Pafnuty Chebyshev, in the mid-nineteenth century.
Uppercase letters such as X or Y denote a random variable. Lowercase letters like x or y denote the value of a random variable. If X is a random variable, then X is written in words, and x is given as a number.
For example, let X = the...
Uppercase letters such as X or Y denote a random variable. Lowercase letters like x or y denote the value of a random variable. If X is a random variable, then X is written in words, and x is given as a number.
For example, let X = the...
17.6K
01:26Variables Affecting Phosphorescence and Fluorescence
1.3K
Fluorescence and phosphorescence are essential phenomena in fields like analytical chemistry, biological imaging, and materials science, where they detect molecular properties and visualize cellular structures. Understanding the variables that influence these luminescent behaviors is crucial for maximizing accuracy and efficiency in their applications. These variables can broadly be grouped into chemical structure, solvent properties, and external conditions, each playing a distinct role in...
1.3K
01:19Normal Stress
1.4K
Normal stress is a type of stress that occurs when forces act perpendicular, or normal, to a material's cross-sectional area. This stress often arises in structures when subjected to axial loading, which is the application of force along the axis of an object. A practical example of this can be found in bridge truss members.
When a rod is under axial loading, the internal forces and corresponding stress are normal to the plane of the section, so it is termed normal stress. It's important to...
When a rod is under axial loading, the internal forces and corresponding stress are normal to the plane of the section, so it is termed normal stress. It's important to...
1.4K