A5032540144- Statistical Distribution Estimation and Applications
- Stochastic processes and financial applications
- Probabilistic and Robust Engineering Design
- Statistical Methods and Inference
- Reliability and Maintenance Optimization
- Nonlinear Differential Equations Analysis
- Stochastic processes and statistical mechanics
- advanced mathematical theories
- Advanced Statistical Methods and Models
- Statistical Methods in Clinical Trials
- Bayesian Methods and Mixture Models
- Statistical Methods and Bayesian Inference
- Software Reliability and Analysis Research
- Forecasting Techniques and Applications
- Financial Risk and Volatility Modeling
- Network Security and Intrusion Detection
- Bayesian Modeling and Causal Inference
- Statistics Education and Methodologies
- Atmospheric and Environmental Gas Dynamics
- Probability and Risk Models
- Advanced Statistical Process Monitoring
- Information and Cyber Security
- Stock Market Forecasting Methods
- Mathematical and Theoretical Epidemiology and Ecology Models
- Mathematical Biology Tumor Growth
University of South Florida
2015-2024
Siena College
2024
University of Michigan–Dearborn
2016
University of Tampa
1976-2014
Hong Kong University of Science and Technology
2011
University of Hong Kong
2011
Radford University
1974-2011
Purdue University Northwest
2009
Sewanee: The University of the South
1979
Iowa Methodist Medical Center
1978
Introduction 1. Simple descriptive methods of analysis 2. Theory stationery processes 3. Spectral 4. Repeated measurements 5. Fitting autoregressive moving average to data 6. Forecasting 7. Elements bivariate time-series References Appendix A, B & C
Imbalanced class distribution is an inherent problem in many real-world classification tasks where the minority of interest. Many conventional statistical and machine learning algorithms are subject to frequency bias, discriminating boundaries between majority classes could be challenging. To address imbalance deep learning, we propose a rebalancing strategy based on class-balanced dynamically weighted loss function weights assigned predicted probability ground-truth class. The ability...
Summary The object of the present study is to summarize recent developments in nonparametric density estimation. covers period time from 1956 1978. Most important types estimations are discussed. These include Parzen or kernel estimators, series penalized maximum likelihood and various other estimation techniques.
Parkinson’s Disease (PD) is a devastating neurodegenerative disorder that affects millions of people around the globe. Many researchers are continuously working to understand PD and develop treatments improve condition patients their day-to-day lives. Since last decades, treatment, Deep Brain Stimulation (DBS) has given promising results for motor symptoms by improving quality daily living patients. In methodology present study, we have utilized sophisticated statistical approaches...
Abstract Pancreatic cancer is one of the deadliest carcinogenic diseases affecting people all over world. The majority patients are usually detected at Stage III or IV, and chances survival very low once late stages. This study focuses on building an efficient data-driven analytical predictive model based associated risk factors identifying most contributing influencing times diagnosed with pancreatic using XGBoost (eXtreme Gradient Boosting) algorithm. grid-search mechanism was implemented...
A non-homogeneous Poisson process has empirically been shown to be useful in tracking the reliability growth of a system as it undergoes development. It is interest estimate failure intensity this model at time n. The maximum likelihood known, but desirable have Bayesian allow for input prior information. Since ordinary Bayes approach appears mathematically intractable, quasi-Bayes taken. proposed qualitative properties one anticipates from estimate, easy compute. numerical example...
In the theory of turbulence, random position a tagged point in continuous fluid turbulent motion, r(t; ω), is vector-valued function time t ≥ 0, ω ε Ω, where Ω supporting set underlying probability space (Ω, B, P). If u(r, t; ω) Eulerian velocity field, then satisfies stochastic integral equation r(t;ω)=∫0tu(r(Υ;ω),Υ;ω)dΥ, t≥0. General conditions under which solution this exists are given form theorem, and theorem proved using concepts admissibility with respect to an operator on Banach...
In this paper a stochastic model for stream pollution is given which involves random differential equation of the form \[( * )\qquad \dot {\bf X}( t ) = A}{\bf + Y},\quad t\geqq 0,\] where ${\bf )$ two-dimensional vector-valued process with first component giving biochemical oxygen demand (BOD) and second representing dissolved (DO) at distance downstream from source pollution. The fundamental Liouville’s theorem utilized to obtain probability distribution solution $( ),{\bf )$, each various...