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 15: Data Synthetic and the Multiverse of Modeling
In this episode, I assessed applications for AI-generated synthetic data, read up on the radical R tutorial rflow, and stepped into the ‘multiverse’ package to manage reporting on many different statistical analyses.
References:
- Machine learning techniques for forecasting agricultural prices: A case of brinjal in Odisha, India
- The risk of bias in denoising methods: Examples from neuroimaging
- Bayesian model averaging for nonparametric discontinuity design
- AI-generated Synthetic Data, easy and fast access to high quality data?
- Use functions in a WHERE statement to filter observations
- Cronbach’s Alpha: Definition, Calculations & Example
- Updated Comparison of R Graphical User Interfaces
- A Major Contribution to Learning R
R-packages:
- multiverse: 'Explorable Multiverse' Data Analysis and Reports
6 min - Episode 14: Outliers from the Reaches of Sample Space
In this episode, I explored model selection methods for fending off invading outliers, marveled at May’s Top 40 CRAN packages, refreshed my suppositions on samples space, and rallied up to replace words in text using package ‘rflashtext’.
References:
- Robust model selection using the out-of-bag bootstrap in linear regression
- Data-informed deep optimization
- Compute the multivariate t density function
- Sample Space Examples and The Counting Principle
- Cohens D: Definition, Using & Examples
- Statistical Inference: Definition, Methods & Example
- Replace NA with Zero in R
- May 2022: “Top 40” New CRAN Packages
R-packages:
- sciRmdTheme: Upgraded 'Rmarkdown' Themes for Scientific Writing
- rflashtext: FlashText Algorithm for Finding and Replacing Words
6 min - Episode 13: Funny-Looking Bioinformatics
In this episode, I broadened up my bioinformatics background, found that the latest R-release was a Funny-Looking Kid, and extracted lobster species using package ‘OpenTreeChronograms’.
References:
- A New Tobit Ridge-Type Estimator of the Censored Regression Model with Multicollinearity Problem
- Selecting predictive biomarkers from genomic data
- COBREXA.jl: constraint-based reconstruction and exascale analysis
- Partial leverage plots
- R 4.2.1 is released
- {CIPerm}: Computationally-Efficient Confidence Intervals for Mean Shift from Permutation Methods
- When Linear Models Don’t Fit Your Data, Now What?
- SVG to PNG
- How to Find Unmatched Records in R
- Create new variables from existing variables in R
R-packages:
- OpenTreeChronograms: Open Tree of Life Chronograms
- phacking: Sensitivity Analysis for p-Hacking in Meta-Analyses
- RCTS: Clustering Time Series While Resisting Outliers
- MakeYourPalette: Make Your Palette
- ontologics: Code-Logics to Handle Ontologies
- CIPerm: Computationally-Efficient Confidence Intervals for Mean Shift from Permutation Methods
8 min - Episode 12: Bayesian Fire Engine
In this episode, I tackled ‘omics’ data with a Bayesian fire engine, soared over a timeline of statistical ideas, and decreased statistical bias by increasing bias understanding.
References:
- A Survey of Statistical Methods for Microbiome Data Analysis
- PyCon US: Successful Return to In-Person in 2022
- PUSH, POP, and reset options for ODS graphics
- Using Bayesian Additive Regression Trees for Flexible Outcome Modeling
- R 4.2.1 scheduled for June 23
- Bias in Statistics: Definition, Selection Bias & Survivorship Bias
- X-MR (X-Moving Range) Chart
- From Anna Menacher: A timeline of the most important statistical ideas of the past 50 years
- useR! 2022 – all virtual – is next week!
R-packages:
- bnClustOmics: Bayesian Network-Based Clustering of Multi-Omics Data
- starticles: A Generic, Publisher-Independent Template for Writing Scientific Documents in 'rmarkdown'
6 min - Episode 11: Putting the Meta in Meta and Drooling Over Donut Plots
In this episode, I learned there are both good and bad R books, umbrella reviews are meta on top of meta, flowcharts can be fashioned in ggplot2, and package ‘ggpie’ is more than mere fluff.
References:
- Monotone Regression: A Simple and Fast O(n) PAVA Implementation
- Pro Data Visualization Using R and JavaScript: Analyze and Visualize Key Data on the Web
- Doing Meta-Analysis with R - A Hands-On Guide
- Convergence analysis and approximate optimal temporal step sizes for some finite difference methods discretising Fisher’s equation
- Early warning signals of financial crises using persistent homology and critical slowing down: Evidence from difference correlation tests
- Methodological approaches for assessing certainty of the evidence in umbrella reviews: A scoping review
- Python 3.10.5 is available
- Weights for residuals in robust regression
- The effect of weight functions in a robust regression method
- Considerations for generating time-accurate animation with SAS
- Extra! Extra! Read all about how SAS is leading in analytics!
- Bug BBQ
- What Is Specification Error in Statistical Models?
- Becoming an R developer: the workshop
- Creating flowcharts with {ggplot2}
R-packages:
- clinicalsignificance: Determine the Clinical Significance in Clinical Trials
- mlrpro: Stepwise Regression with Assumptions Checking
- ggpie: Pie, Donut and Rose Pie Plots
- gTestsMulti: New Graph-Based Multi-Sample Tests
- plnr: A System for Planing Analyses
- holiglm: Holistic Generalized Linear Models
9 min - Episode 10: Multiple Last Names and Growing a Backbone
In this episode, I chained together multiple scientific last names, found heaps of PLOS ONE journal articles, and learned how to grow a backbone in R using package ‘backbone’.
References:
- Event History Regression with Pseudo-Observations: Computational Approaches and an Implementation in R
- More on Multidimensional Scaling and Unfolding in R: smacof Version 2
- A Practical Guide to the Numerical Implementation of Tensor Networks I: Contractions, Decompositions, and Gauge Freedom
- Rank-polyserial correlation. Quest for a "missing" coefficient of correlation
- Assessing and visualizing fragility of clinical results with binary outcomes in R using the fragility package
- backbone: An R package to extract network backbones
- Evaluation of spatial Bayesian Empirical Likelihood models in analysis of small area data
- Clinical artificial intelligence quality improvement: towards continual monitoring and updating of AI algorithms in healthcare
- The interplay between ranking and communities in networks
- Random assignment of subjects to groups in SAS
- Using formatted CLASS variables
- Python 3.11.0b2 is now available
- shadowr: Selenium plugin to manage multiple levels of shadow elements on web page
- Difference Between Standard Deviation and Standard Error
- Clopper-Pearson Exact Method
- Applications of Python | Top 10 uses of Python for The Real World
- 15 Fun & Easy Excel Project Ideas To Sharpen Your Skills
- 2022-01 Vectorised Pattern Fills in R Graphics
- April: “Top 40” New CRAN Packages
- R Packages Download Stats
R-packages:
- AHSurv: Flexible Parametric Accelerated Hazards Models
- misPRIME: Partial Replacement Imputation Estimation for Missing Covariates
- tracee: Easily Save Output and Trace it Back to Code
- backbone: Extracts the Backbone from Graphs
10 min - Episode 9: Boxes and Babies
In this episode I thought outside the Box-Pierce; fuzzed, sweated, and flustered my way through basic stats; encountered functions and AI for babies; and chronicled the R-package ‘chronicler’.
References:
- NeuralSens: Sensitivity Analysis of Neural Networks
- econet: An R Package for Parameter-Dependent Network Centrality Measures
- Modified Quantile Regression For Modeling the Low Birth Rate
- A Novel Correction for the Adjusted Box-Pierce Test
- Comments on identifying causal relationships in nonlinear dynamical systems via empirical mode decomposition
- Adaptive numerical simulations with Trixi.jl: A case study of Julia for scientific computing
- The balanced bootstrap in SAS
- Game-changer AI tool will save mothers and babies
- D. Mayo & D. Hand: “Statistical significance and its critics: practicing damaging science, or damaging scientific practice?”
- Fuzzy Clustering: Definition
- How to Find the P value: Process and Calculations
- The Difference Between an Odds Ratio and a Predicted Odds
- ‘Data analysis with tidyverse’ workshop
- Think like a programmeR: the workshop
- How to add labels at the end of each line in ggplot2?
- Subsetting with multiple conditions in R
- Hierarchical data visualization with Shiny and D3
- chronicler: Add Logging to Functions
R-packages:
- Rwclust: Random Walk Clustering on Weighted Graphs
- ScaleSpikeSlab: Scalable Spike-and-Slab
- webshot2: Take Screenshots of Web Pages
- mverse: Tidy Multiverse Analysis Made Simple
- chronicler: Adding Logging to Functions
11 min - Episode 8: Shortened Blankets
In this episode, I got lost in among the statistical trees, flipped a Bernoulli trial coin, and wrapped myself up in package ‘ggblanket’.
References:
- sensobol: An R Package to Compute Variance-Based Sensitivity Indices
- The R Package stagedtrees for Structural Learning of Stratified Staged Trees
- Wave Effects of the Fractional Shallow Water Equation and the Fractional Optical Fiber Equation
- Two Domains of Meandering Spiral Waves in a Modified Barkley Model
- Access for success: SAS Certification practice exams now offered for free
- How to unroll frequency data
- Sampling Methods: Different Types in Research
- Bernoulli Trials: Definition, Examples
- ANCOVA: Analysis of Covariance
- Measures of Model Fit for Linear Regression Models
- ggblanket: making beautiful ggplot2 visualisation simpler
- Self-documenting {ggplot}s thanks to the power of monads!
- Launch of R Screencasts
- How to draw heatmap in r: Quick and Easy way
R packages:
- DemographicTable: Creating Demographic Table
- Linea: Linear Regression Interface
- linelist : Tagging and Validating Epidemiological Data
- NovelDistns: Computes PDF, CDF, Quantile, Random Numbers and Measures of Inference for 3 General Families of Distributions
- semtree: Recursive Partitioning for Structural Equation Models
- getwiki R: Wrapper for Wikipedia Data
- simer: Data Simulation for Life Science and Breeding
- ggblanket: Wrappers to Make Beautiful 'ggplot2' Visualization Simpler
- dimensionalAnalysis: Dimensional Analysis
9 min - Episode 7: Riding the Python Lightning
In this episode, I dived into non-normal distributions, went from rags to ridges, rode the Python Language Summit lightning, and looked at the ‘GenericML’ package.
References:
- Covsim: An R package for Simulating Non-Normal Data for Structural Equation Models Using Copulas
- Rags2ridges: A One-stop-L2-shop for Graphical Modeling of High-Dimensional Precision Matrices
- Transmissibility in Interactive Nanocomposite Diffusion: The Nonlinear Double-Diffusion Model
- Editorial: Modelling collective motion across scales
- The Application of the Functional Variable Method for Solving the Loaded Nonlinear Evaluation Equations
- The derivative of the determinant of a matrix
- Python 3.11.0b1 is now available
- The 2022 Python Language Summit: Upstreaming optimisations from Cinder
- The 2022 Python Language Summit: Performance Improvements by the Faster CPython team
- The 2022 Python Language Summit: Python in the browser
- The 2022 Python Language Summit: A per-interpreter GIL
- The 2022 Python Language Summit: Achieving immortality
- Fiducial Inference: Fisher’s Great Failure
- Beta Distribution: Uses, Parameters & Examples
- Geometric Distribution: Uses, Calculator & Formula
- Alignment cheatsheet
- 5 Key Data Visualization Principles Explained – Examples in R
- Three packages that port the tidyverse to Python
- How to embed a Shiny app into your blog posts
R packages:
- biogeom: Biological Geometries
- GenericML: Generic Machine Learning Inference
- PMA2: Penalized Multivariate Analysis
9 min - Episode 6: Mixes and Metabolomics
In this episode, I dig into the first articles of a new Journal of Statistical Software volume, try not to mix up package ‘nvmix’ with ‘nmix’, scratch my head over Navier-Stokes equations, and take the random R-package section in a metabolomics direction.
References:
- Journal of Statistical Software Volume 102
- Convolutional Neural Networks for Very Low-Dimensional LPV Approximations of Incompressible Navier-Stokes Equations
- Simulate the null distribution for a hypothesis test
- Nmix package - Bayesian inference about univariate Gaussian mixtures with an unknown number of components
- What is Power in Statistics?
- Member Training: Introduction to SPSS Software Tutorial
- Announcing R/Medicine 2022!
- New Workshops Series Kick-off
- Little useless-useful R functions – Animating datasets
- March: “Top 40” New CRAN Packages
- An introduction to the metabolomicsR
R packages:
- arthistory: Art History Textbook Data
- nls2: Non-Linear Regression with Brute Force
- metabolomicsR: Tools for Metabolomics Data
- conductor: Create Tours in 'Shiny' Apps Using 'Shepherd.js'
10 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.