About the service
My main tool for statistics and data analysis is the R language, and over years of work I have learned to turn chaotic tables and raw exports into results worth defending. I do real statistics, not general scripting: I fit and check regressions, mixed models, GLM, survival and time series models, run hypothesis tests and power calculations, plan samples and carefully verify assumptions so conclusions hold up to scrutiny. If you have a research question, a dataset, and a deadline, I will take the task from raw file to a clear and honest answer.
In daily work I live inside tidyverse. I clean, transform, and join data with dplyr and tidyr, automate repeating pipelines, and document every transformation so the entire analysis runs with one command. For clarity, I build publication-quality graphics in ggplot2 — from exploratory plots to polished illustrations for papers and dashboards, and I format results as R Markdown or Quarto reports that build straight into HTML, PDF, or Word. This clearly sets me apart from Python data scientists, Julia developers, and BI dashboard specialists who drag blocks around: my entire toolkit is the R statistical ecosystem.
When the result needs to be interactive rather than a static report, I develop Shiny applications so your team can explore data, filter it, and rerun models without a single line of code. I value correctness, transparent methodology, and readable code, and I explain the meaning of numbers in plain language including caveats and limits. A one-off analysis, academic research, an A/B test, or a standing reporting pipeline — I deliver work you can trust and reuse.
— Statistical modeling: regressions, GLM, mixed models, survival, time series
— Hypothesis testing, power analysis, and experiment design
— Data processing in tidyverse: dplyr, tidyr, purrr
— Data visualization in ggplot2 for papers and dashboards
— Interactive web apps in Shiny
— Reproducible reports in R Markdown and Quarto
In daily work I live inside tidyverse. I clean, transform, and join data with dplyr and tidyr, automate repeating pipelines, and document every transformation so the entire analysis runs with one command. For clarity, I build publication-quality graphics in ggplot2 — from exploratory plots to polished illustrations for papers and dashboards, and I format results as R Markdown or Quarto reports that build straight into HTML, PDF, or Word. This clearly sets me apart from Python data scientists, Julia developers, and BI dashboard specialists who drag blocks around: my entire toolkit is the R statistical ecosystem.
When the result needs to be interactive rather than a static report, I develop Shiny applications so your team can explore data, filter it, and rerun models without a single line of code. I value correctness, transparent methodology, and readable code, and I explain the meaning of numbers in plain language including caveats and limits. A one-off analysis, academic research, an A/B test, or a standing reporting pipeline — I deliver work you can trust and reuse.
— Statistical modeling: regressions, GLM, mixed models, survival, time series
— Hypothesis testing, power analysis, and experiment design
— Data processing in tidyverse: dplyr, tidyr, purrr
— Data visualization in ggplot2 for papers and dashboards
— Interactive web apps in Shiny
— Reproducible reports in R Markdown and Quarto
