mylibrary

This repository centralizes rendered examples for data analysis, statistics, bibliographic workflows, interoperability between languages, and small applied experiments. The source notebooks live under Rmd/; the index below points only to the generated Markdown files in examples/.

Example Index

• Complete Index of Examples
• Seed Utility README

BigQuery

• BigQuery Walkthrough - Query BigQuery tables, join geographic data, and derive simple population-based estimates.

Mobility

• Merge and Save Mobility Data - Combine partitioned mobility data into a single output artifact.

PyTorch

• Generate Classification Data - Build a synthetic dataset for tabular classification experiments.
• Generate Regression Data - Build a synthetic dataset for regression experiments.
• Generate Time Series Data - Create univariate time series data for forecasting tests.
• Generate Time Series Classification Data - Create labeled temporal sequences for classification tasks.
• Iris Classification in R - Train and inspect a simple classification workflow using the Iris dataset.
• Event Classification - Prepare and model event-oriented data for supervised classification.
• Autoencoder Example - Train an autoencoder and inspect compressed representations or reconstruction behavior.

R

• CRAN Downloads and Country Distribution - Analyze CRAN downloads over time and inspect country distribution for a selected day.

Seed

• Seed Utility README - Documentacao do modulo seed.py para reproducibilidade entre Python, NumPy, PyTorch e R via reticulate.
• Python Seed Example - Exemplo Python chamando seed_everything(123).
• R and Python Seed Example - Exemplo R definindo set.seed(123) e depois o seed do Python.

References

• Reference Utilities Overview - Overview of the bibliography and LaTeX utilities available in this repository.
• Build Query Strings from BibTeX - Generate boolean search expressions from DOI fields or normalized titles.
• Check BibTeX File for Issues - Validate entries and report missing fields or common inconsistencies.
• Clean a Single BibTeX File - Normalize one bibliography file before reuse.
• Clean All BibTeX Files in a Directory - Apply the cleaning workflow to every .bib file in a directory tree.
• Expand LaTeX Includes - Resolve \input{}, \include{}, \import{}, and related directives into a single expanded file.
• Export Google Scholar Publications - Collect an author's publications and export the result to a spreadsheet.
• Find and Remove Unused References in a Directory - Detect unused citation keys across multiple .tex files and prune the bibliography.
• Find and Remove Unused References in a Single File - Detect unused citation keys for one main LaTeX document and prune the bibliography.
• Map and Replace Reference Keys in a Directory - Update BibTeX keys and propagate the replacements across a directory of LaTeX files.
• Map and Replace Reference Keys in a Single File - Update BibTeX keys and rewrite citations in one LaTeX document.
• Merge BibTeX Files Pairwise - Reconcile multiple .bib files through pairwise merges.
• Print DOI URLs - Emit https://doi.org/... links for entries that contain DOI values.
• Union of BibTeX Files - Build a consolidated bibliography from a directory of source files.

Reticulate

• Invoke Python from R with Reticulate - Exchange data and function calls between R and Python.

Statistics

• Regression - Explore linear, polynomial, multiple, and logistic regression with plots and worked examples.
• Effect Size Tutorial - Compare paired methods and compute effect size alongside nonparametric testing.
• Nonparametric Tests Tutorial - Perform normality checks and Wilcoxon-based comparisons for independent and paired samples.
• Parametric Tests Tutorial - Review t-test variants, assumptions, and interpretation for common experimental setups.

Stocks

• IBX50 Example - Load and explore the IBX50 stock dataset with a small analysis workflow.

Time Series

• Reshape Data with Pivot Operations - Transform time-indexed tables between wide and long formats.
• Forecasting with ARIMA and Fourier Terms - Fit a seasonal forecasting model using ARIMA components and Fourier features.
• Kalman Filter for AR(1) - Implement and visualize a Kalman filtering workflow for an autoregressive process.
• Event Detection with K-Means - Cluster temporal behavior to identify event-like patterns.
• Multivariate Anomaly Detection with PCA - Use principal components to monitor multivariate trajectories and flag anomalies.
• Animated Detections - Build animated visual summaries for evolving detections over time.
• Maximum Likelihood for the Binomial Model - Derive and inspect likelihood and log-likelihood behavior for a binomial parameter.
• Change Detection with EWMA - Use exponentially weighted moving averages to detect shifts in a process.
• MLE for the Normal Mean - Estimate the mean parameter by maximizing the normal log-likelihood.
• MLE for Normal Mean and Variance - Estimate multiple normal parameters under a constrained optimization setup.
• Density-Based Outlier Detection - Score observations with density estimation and highlight low-probability points.
• Entropy and Surprisal for Anomaly Detection - Convert empirical probabilities into surprisal-based anomaly scores.
• Rolling Grubbs Test for Anomaly Detection - Apply a rolling hypothesis test to identify local outliers.
• Plot System Fonts - Enumerate and visualize fonts available in the local plotting environment.
• Color Palette Demo - Compare palette choices for charts and exploratory graphics.
• Spline Example - Fit and visualize spline-based curves for smooth approximation.

TSFM

• Time Series Foundation Model Example - Run a compact end-to-end example with a toy time series foundation model.

Tutorial

• Additional Graphics Examples - Small plotting demonstrations complementary to the statistics material.
• Plot Drift Example - Illustrate how distribution drift can be visualized over time.