Haibiostat
Haibiostat
Recently Published
Skin Tone Scale
Three different skin tone color scale applied in social/racial research
PGM - Factors
Factors - Why needed them?
PGM - Distributions
Distributions: Joint - Conditioning - Marginalization
Getting the individual probability from Gradient Boosting Models
Goal:
+ Fitting logistic model vs GBM
+ Extract the individual probability from GBM
+ Calculating the accuracy between logistic vs GBM
+ Comparing individual probability between 2 methods
ReadMe --- HaiBiostats' Portfolio
"Entertainment" Projects' Outline
Fundamentals of Bayes' Theorem
Demonstrate the degree of belief (prior) interfered/updated the evidence (likelihood) to produce a piece of new evidence (posterior)
Variance, Standard Deviation and Standard Error Confusion
Explanation of Var, SD and SEM
Creating model diagrams in JAGS/Stan model String process: a tool to draw
- Introducing an useful tool to draw diagrams
Scaling Back the Standardized Betas in JAGS
Converting Standardized Beta Coefficient Estimates to Raw Data Scale
Bayes' Analysis 7 - Part 2
- Simple linear regression
- Hierarchical regression
Bayes' Analysis 8 - Part 1
- Multiple regression
When fitting random-intercept vs. random slope mixed-effect regression model
- Question
- Review
- Answer
Bayes' Analysis 7 - Part 1
- Compare 2 groups without predictors
- FNP vs Bayes in comparison of groups
ANOVA in Bayes
ANOVA one-way, two-way
Bayes' Analysis 6
- Normal model for 1 group
- Robust estimation using t-distribution
Bayes' Analysis 5
Introducing to STAN
Beta distribution in an intuitive explanation
- Describe beta dist. in an intuitive way
- With some math formulas
Interaction - Power analyses
- Explain the meaning of interaction
- Power analysis to detect an interaction effect
- Simulation of the power
COVID-19 Vaccine efficacy
VE and 95% CI of VE
Cox Models - PH assumption - ML applied in SA
KM - Cox - PH assumption - Machine learning in Survival Analysis
Three Phases in Clinical Trials
A short review of Phases in Clinical Trials - easier to remember
stringi vs. stringr
How stringi evolved from stringr!
Gentle Introduction to Shiny
Basic Steps in Create an App with Shiny
Functions inside a function
Review a function structure - An applicant
Centile Estimation using GAMLSS
Centile estimation:
- Intro
- Fitting centile curves, Plotting, Diagnostics
- Prediction, Quantile sheets
Regular Expressions applying in the Text Mining
- Review regular expression and functions in R dealing with Text Mining
- Demonstrating the applications
The Evolution of Reshaping Data
From stats::reshape -- all kinds of tidyr function (gather, spread, separate, unite)
Introduction to Bootstrap and Some Examples
Bootstrap
Examples
ML - Probabilistic Model Selection with AIC, BIC
- The Challenge of Model Selection
- Probabilistic Model Selection
- Akaike Information Criterion
- Bayesian Information Criterion
- Minimum Description Length
- Examples
Probabilistic Graphical Models - Template Models for Bayesian Networks
Overview - Temporal Models (DBNs, HMMs) - Plate Models
Probabilistic Graphical Models - Bayesian Network: Fundamentals
Bayesian networks build on the same intuitions as the naive Bayes model by exploiting conditional independence properties of the distribution in order to allow a compact and natural representation. We look at the fundamental of BNs
ML - Classifying with Trees -- Decision Trees
What are decision trees?
Using the recursive partitioning algorithm to predict animal classes
An important weakness of decision trees
Adjustment for Baseline Response in Longitudinal Analysis
- Review 4/6 strategies in baseline response adjustment
Probability of Distribution in R
- Some distributions often meet in R
- Data.Frame example
ML - Factor Analysis applied in Multiple Linear Regression Context using R
- Scree plot in PCA and FA
- Run fa and factanal
- Applying fa$scores in mutiple regression analysis, dimension reduction process.