Sign up to save your podcasts
Or
Share Science Research Weekly
Share to email
Share to Facebook
Share to X
Share to email
Share to Facebook
Share to X
Or
Science Research Weekly
By Mark R WilliamsonScience
Science Research Weekly episodes
- Episode 25: Machine Learning vs. Magic
In this episode, I found that Bambi (Bayesian Model Building Interface) is more than a little deer, rooted for the round-robin operator for genetic algorithms, and had my convictions that machine learning is not magic reinforced.
References:
- HighFrequencyCovariance: A Julia Package for Estimating Covariance Matrices Using High Frequency Financial Data
- Bambi: A Simple Interface for Fitting Bayesian Linear Models in Python
- Spbsampling: An R Package for Spatially Balanced Sampling
- plot3logit: Ternary Plots for Interpreting Trinomial Regression Models
- Learning Base R (2nd Edition)
- Python and R for the Modern Data Scientist
- Genetic algorithm with a new round-robin based tournament selection: Statistical properties analysis
- Healthcare researchers must be wary of misusing AI
- New method to identify symmetries in data using Bayesian statistics
- Python 3.11.0rc2 is now available
- Complex Layouts using the SG Procedures
- Factor Analysis Guide with an Example
- How to Choose Appropriate Clustering Method for Your Dataset
- How to Apply AI to Small Data Sets?
- The R Consortium Needs Your Help with satRdays
- Visualizing OLS Linear Regression Assumptions in R
R-packages:
- TPCselect: Variable Selection via Threshold Partial Correlation
- historicalborrow: Non-Longitudinal Bayesian Historical Borrowing Models
- historicalborrowlong: Longitudinal Bayesian Historical Borrowing Models
- kgp: 1000 Genomes Project Metadata
- DBIsqldf: Manipulate R Data Frames Using SQL
- latentFactoR: Data Simulation Based on Latent Factors
- TSdeeplearning: Deep Learning Model for Time Series Forecasting
9 min - Episode 24: The Way of the R Warrior
In this episode, I chose mediation analysis over a hospital heist, gained additional sample size calculation insights, memorized machine learning methods, and battled my way up a text-based tower using package ‘rwarrior’.
References:
- Hierarchical Clustering with Contiguity Constraint in R
- Blang: Bayesian Declarative Modeling of General Data Structures and Inference via Algorithms Based on Distribution Continua
- Robust Mediation Analysis: The R Package robmed
- A Software Tool For Sparse Estimation Of A General Class Of High-dimensional GLMs
- bayesanova: An R package for Bayesian Inference in the Analysis of Variance via Markov Chain Monte Carlo in Gaussian Mixture Models
- blindrecalc - An R Package for Blinded Sample Size Recalculation
- Power and Sample Size for Longitudinal Models in R – The longpower Package and Shiny App
- RFpredInterval: An R Package for Prediction Intervals with Random Forests and Boosted Forests
- Alternative stopping rules to limit tree expansion for random forest models
- gMCPLite: Lightweight Graph Based Multiple Comparison Procedures
- pkglite 0.2.1 is released
- Python releases 3.10.7, 3.9.14, 3.8.14, and 3.7.14 are now available
- Ordinal Data: Definition, Examples & Analysis
- How the Population Distribution Influences the Confidence Interval
- Talks to watch at the RSS International Conference
- How to Avoid Overfitting?
- Beneath and Beyond the Cox Model (www.r-bloggers.com/2022/09/beneath-and-beyond-the-cox-model/)
- What are the algorithms used in machine learning? (www.r-bloggers.com/2022/09/what-are-the-algorithms-used-in-machine-learning/)
- 23 New books added to Big Book of R (www.r-bloggers.com/2022/09/23-new-books-added-to-big-book-of-r/)
- R Warrior Leader Board (www.r-bloggers.com/2022/08/r-warrior-leader-board/)
R-packages:
- oceanexplorer: Explore Our Planet's Oceans with NOAA
- OneArm2stage: Optimal One-Arm Two-Stage Phase II Design with Survival Endpoint
- rwarrior: R Warrior - An AI Programming Game
- manymome: Mediation, Moderation and Moderated-Mediation After Model Fitting
- voice: Tools for Voice Analysis, Speaker Recognition and Mood Inference
- shinylight: Web Interface to 'R' Functions
11 min - Episode 23: Biting Off More Statistics Than I Can Chew
In this episode, I bite off more journal articles than I could chew, bookmarked how to reorder SGPLOT bars, was underwhelmed by R surprises, and muddled through the meta-package ‘easystats’.
References:
- On the Programmatic Generation of Reproducible Documents
- Automatic Identification and Forecasting of Structural Unobserved Components Models with UComp
- exuber: Recursive Right-Tailed Unit Root Testing with R
- The R Journal Volume 14/1
- A unified Framework for analyzing complex systems: Juxtaposing the (Kernel) PCA method and Graph Theory
- Grand Challenges in Mathematical Biology: Integrating Multiscale Modelling and Data
- CohensdpLibrary: Compute Cohen's d and its confidence interval in any experimental design
- Two types of syntax for the SELECT-WHEN statement in SAS
- Order the bars in a bar chart with PROC SGPLOT
- Top 10 online data science programs
- How to handle Imbalanced Data?
- July 2022: “Top 40” New CRAN Packages
- Surprising Things You Can Do With R
- SAS vs R Programming: Which to Choose and How to Switch
R-packages:
- easystats: Framework for Easy Statistical Modeling, Visualization, and Reporting
- once: Execute Expensive Operations Only Once
- snSMART: Small N Sequential Multiple Assignment Randomized Trial Methods
- tidyplus: Additional 'tidyverse' Functions
- transPlotR: Visualize Transcript Structures in Elegant Way
12 min - Episode 22: Python with a Side of Waffles
In this episode, I grabbed the tools to conduct mobility data analysis and interactive plots in Python, detected outliers in spherical data, considered the next programing language to learn, and readied the statistical syrup for waffle plots using package ‘baffle’.
References:
- scikit-mobility: A Python Library for the Analysis, Generation, and Risk Assessment of Mobility Data
- spNNGP R Package for Nearest Neighbor Gaussian Process Models
- Feller-Pareto and Related Distributions: Numerical Implementation and Actuarial Applications
- A new outlier detection method for spherical data
- Confidence bands in survival analysis
- How to stagger labels on an axis in PROC SGPLOT
- Experimental Design: Definition and Types
- The Difference Between R-squared and Adjusted R-squared
- Top 5 Python Libraries for Data Science (2023) Edition
- Base-R and Tidyverse Code, Side-by-Side
- R Design Patterns, Base-R vs. Tidyverse With a view toward the teaching of R beginners
- Hello Shiny Python
- Which programming language should I learn?
- R Markdown Tips: Code, Images, Comments, Tables, and more
- Is Data Science a Dying Profession?
R-packages:
- evalR Evaluation of Unverified Code
- goldfish Statistical Network Models for Dynamic Network Data
- baffle Make Waffle Plots with Base Graphics
- ggDoE Modern Graphs for Design of Experiments with 'ggplot2'
- jjAnno An Annotation Package for 'ggplot2' Output
- multinomialLogitMix Clustering Multinomial Count Data under the Presence of Covariates
- IntLIM Integration of Omics Data Using Linear Modeling
- scrutiny Error Detection in Science
- shinyHugePlot Efficient Plotting of Large-Sized Data
- ctmva Continuous-Time Multivariate Analysis
- ggcoverage Visualize Genome Coverage with Various Annotations
8 min - Episode 21: Behind the Statistical Software Curve
In this episode, I played catch-up with the Journal of Statistical Software, schemed how to snag a top spot in the XP Learning Challenge, and roared alongside the R-package ‘TRexSelector’.
References:
- modelsummary: Data and Model Summaries in R
- stringi: Fast and Portable Character String Processing in R
- evgam: An R Package for Generalized Additive Extreme Value Models
- Julia 1.8 Highlights
- The Box-Cox transformation for a dependent variable in a regression
- TDApplied: Machine Learning and Inference for Topological Data Analysis
- Some project opportunities for Ph.D. students!
- Learn data skills, earn points and win money prizes daily with Datacamp!
- Designing #TidyTuesday visualisations for mobile (with Quarto)
- Find out which data skills are most in demand?
R-packages:
- ctmva: Continuous-Time Multivariate Analysis
- ggcoverage: Visualize Genome Coverage with Various Annotations
- TRexSelector: T-Rex Selector: High-Dimensional Variable Selection FDR Control
- countdown: A Countdown Timer for HTML Presentations, Documents, and Web Apps
- TDApplied: Machine Learning and Inference for Topological Data Analysis
- brassica: 1970s BASIC Interpreter
9 min - Episode 20: The Count of Monte Carlo
In this episode, I browsed yet more Monte Carlo simulation papers, introduced myself to various structural equation models, championed base R, and went neural to neural with the R package ‘cito’.
References:
- A kernel mixing strategy for use in adaptive Markov chain Monte Carlo and stochastic optimization contexts
- Bootstrap-based inferential improvements to the simplex nonlinear regression model
- Python 3.10.6 is available
- Stanine Score: Definition, Examples, How to Convert
- The Four Models You Meet in Structural Equation Modeling
- Downstream Bioinformatics Analysis of Omics Data with edgeR
- Simulating data from a non-linear function by specifying a handful of points
- RObservations #36: Opinions on RStudio’s name change. A Bayesian approach with Stan
- Base-R Is Alive and Well
R-packages:
- cito Building and Training Neural Networks
- cbioportalR Browse and Query Clinical and Genomic Data from cBioPortal
- gtreg Regulatory Tables for Clinical Research
- diffdfs Compute the Difference Between Data Frames
- fake Flexible Data Simulation Using the Multivariate Normal Distribution
- seeker Simplified Fetching and Processing of Microarray and RNA-Seq Data
- ympes Collection of Helper Functions
6 min - Episode 19: Get Python Not Sleep
In this episode, I didn’t get a lot of sleep. But I did get a lot of quirky Python, classy SAS, and ridiculous R.
References:
- A systematic review of statistical methodology used to evaluate progression of chronic kidney disease using electronic healthcare records
- Statistical inference for a constant-stress partially accelerated life tests based on progressively hybrid censored samples from inverted Kumaraswamy distribution
- Python 3.11.0b5 is now available
- How to reveal new connections in a knowledge graph with link prediction
- https://blogs.sas.com/content/iml/2022/08/01/examples-monte-carlo-simulation.html
- 5 pillars for a successful DataOps strategy
- June 2022: “Top 40” New CRAN Packages
- Bioinformatics Analysis of Omics Data with the Shell & R
- Script monkey!
- Getting fun with ggdogs, ggcats and gganimate
R-packages:
- gganimate
5 min - Episode 18: The Software Formerly Known as RStudio
In this episode, RStudio gets a Posit-tively new name, SAS estimates area with Monte Carlo simulation, and I take in the view of package ‘P2C2M.Skyline.’
References:
- Progressive censoring schemes for marshall-olkin pareto distribution with applications
- Assessing model adequacy for Bayesian Skyline plots using posterior predictive simulation
- Testing for Benford’s Law in very small samples: Simulation study and a new test proposal
- Inferring the underlying multivariate structure from bivariate networks with highly correlated nodes
- R Can Use Your Help: Translating R Messages
- Monte Carlo estimates of area
- effectsize: A new Python package for computing effect sizes / standardized differences
- Hoeffding’s Inequality
- Exogenous and Endogenous Variables in Structural Equation Modeling
- Announcing Quarto, a new scientific and technical publishing system
- RStudio is becoming Posit
- Posit– Why Rstudio is changing its name
- How to Create an Interaction Plot in R?
- How to Standardize Data in R?
- R code snippet : Transform from long format to wide format
- Programming a simple minimax chess engine in R
R-packages:
- P2C2M_Skyline
8 min - Episode 17: Tidal Waves of R Packages
In this episode, I pared down three important -omics data considerations, coded my way into wave of new R packages, boasted about the Big Book of R, and havered over the ‘dots’ package.
References:
- Editorial: Statistical Data Science - Theory and Applications in Analyzing Omics Data
- Introducing riskCommunicator: An R package to obtain interpretable effect estimates for public health
- On the mixed Kibria–Lukman estimator for the linear regression model
- Rejection Region (Critical Region) for Statistical Tests
- Join rstudio::conf(2022) Virtually
- ggdensity: A new R package for plotting high-density regions
- R Color
- Generalized Linear Models, Part I: The Logistic Model
- Part 1 of 3: 300+ milestone for Big Book of R
- Part 2 of 3: 300+ milestone for Big Book of R
- Part 3 of 3: 300+ milestone for Big Book of R
R-packages:
- htmcglm: Hypothesis Testing for McGLMs
- nextGenShinyApps: Advanced Tools for Building the Next Generation of 'Shiny' Applications and Dashboards
- robustmeta: Robust Inference for Meta-Analysis with Influential Outlying Studies
- lnmCluster: Perform Logistic Normal Multinomial Clustering for Microbiome Compositional Data
- binaryTimeSeries: Analyzes a Binary Variable During a Time Series
- codebookr: Create Codebooks from Data Frames
- dots: Dot Density Maps
- ggpackets: Package Plot Layers for Easier Portability and Modularization
- trouBBlme4SolveR: Troubles Solver for 'lme4'
7 min - Episode 16: The Wizard of Stats
In this episode, I clued in on the claims that Linear Programming Discriminant analysis is effective, mused on the methods of PROC SQL in SAS, and fired up a virtual machine to use for package ‘vmr’.
References:
- moreThanANOVA: A user-friendly Shiny/R application for exploring and comparing data with interactive visualization
- LPDA: A new classification method based on linear programming
- Speedups in operations with regular expressions
- Why learn SQL?
- Correlated Errors in Confirmatory Factor Analysis
- Network Graphs in R
- vmrPoster
R-packages:
- htmldf: Simple Scraping and Tidy Webpage Summaries
- gpboost: Combining Tree-Boosting with Gaussian Process and Mixed Effects Models
- vmr: Virtual Machines for R
5 min
About Science Research Weekly
From the publisher's feed
A weekly overview of helpful scientific research topics and the successor of the podcast Statistics Weekly. Every week, I will cover 3 topic areas: Research Articles, Research Tools, and Research…
Funding. Science On.