R-MORIE (Multi-domain Open Research and Inferential Estimation) is an R package for causal inference, sampling, psychometrics, point-process modeling, and criminological accountability analysis, with no Python dependencies.
Applied observational research on Canadian carceral, policing, and
oversight data usually means stitching together a dozen single-purpose
packages — one for difference-in-differences, one for propensity-score
matching, one for spatial scan statistics, one for self-exciting point
processes — each with its own data contract, and none aware of the
survey-weighting, provenance, and privacy constraints these data carry.
rmorie is for criminologists, quantitative social scientists, and
accountability researchers who need those estimators in one consistent,
provenance-preserving toolkit, with the MRM (Multilevel Reconciliation
Methodology) framework as its motivating application. Every public
function is prefixed morie_* so it composes safely alongside the
specialist packages it wraps, and results carry the labeling that keeps
synthetic development runs from being mistaken for inferential findings.
Primary methodological references for each estimator family are listed
in the package-level help (?rmorie).
- Reference manual (all
morie_*functions, one searchable doc): https://rootcoder007.r-universe.dev/rmorie/doc/manual.html - Package website (browsable, with articles): https://rootcoder007.github.io/rmorie/
- r-universe project page: https://rootcoder007.r-universe.dev/rmorie
With over 1,900 exported functions, the full reference is large — use the manual or the package site above rather than scrolling the function index. This README covers install + the most common workflows only.
- Over 1,900 exported
morie_*R functions (1,963 at the time of writing) — every public callable is now prefixed to avoid name collisions with other CRAN packages (morie_chi_square_test,morie_kmeans_clustering,morie_decision_tree_split, etc.). The companionmorie.fnPython library mirrors these for cross-language parity. Two deliberate exceptions keep their unprefixed names to match the MRM papers and the Python implementation exactly:mrm_otis_mandela_spectrum()andmrm_classify_mandela(). - SIU subsystem — a full pipeline for the Ontario Special Investigations Unit director's-report corpus (English + French, 2005-present). See SIU pipeline below.
- Free-first AI helpers — local Ollama by default
(
gemma3:4b,translategemma:latest), with optional Gemini, Claude, or Vertex AI fallback. No paid API key is required for the default workflow. - Polite-by-default HTTP fetcher — token-bucket throttling at 4 req/s, exponential backoff on 429/5xx, on-disk page cache.
- Built-in datasets — 41 datasets accessible through the shared
SQLite store (
morie_datasets.db), plus the SIU manifest (4,743 drids, 2,218 unique cases, language-classified). - CPADS contract helpers and IPW / eBAC workflow functions.
- Outputs-manifest tooling — read, validate, audit, and build
outputs_manifest.csvtables for reproducible research projects. - Synthetic data generators for development and CI.
- C/C++ computational backend — Hawkes self-exciting point process
likelihood (Markovian + non-Markovian), HTML-to-text state machine,
SIU parser. See
src/. - Causal-taphonomy suite — Bayesian hierarchical preservation model (cmdstanr / brms / rstanarm HMC backends), absorbing-DTMC decay chains, forensic likelihood ratios, pXRF compositional transforms, and USGS NGDB / MorphoSource open-data ingest.
agent()— call the rmorie CLI agent from R (withagent_available()to probe for the binary).
- Synthetic data is for development, testing, demos, and CI only.
- Final inferential or policy-facing results must be produced from approved real data with full provenance.
- Synthetic runs must be explicitly labeled as synthetic in outputs and reporting text.
From local source:
install.packages("r-package/morie", repos = NULL, type = "source")From r-universe (development snapshot):
install.packages(
"rmorie",
repos = c(rootcoder007 = "https://rootcoder007.r-universe.dev",
CRAN = "https://cloud.r-project.org")
)
# want every optional package too, in one shot? add dependencies = TRUE
# (large: compiles many specialist packages — see "Optional packages" below)
install.packages(
"rmorie", dependencies = TRUE,
repos = c(rootcoder007 = "https://rootcoder007.r-universe.dev",
CRAN = "https://cloud.r-project.org")
)The assistant bridge supports a local fallback through the Python package when no live OpenAI / Anthropic credentials are configured.
The base install stays deliberately light: rmorie wraps many
specialist CRAN packages and declares them in Suggests, so nothing
heavy compiles until you need it. Every function that uses one tells
you exactly what to install when it's missing, and the test suite
skips (never fails) without them. To provision up front instead:
# install every optional package rmorie can use (one-time, ~15 min)
morie_install_extras(which = "all", ask = FALSE)
# or just what's missing, interactively
morie_install_extras()
# or a specific family, e.g. machine learning
morie_install_extras(which = c("randomForest", "glmnet", "xgboost",
"ranger", "caret", "pROC"))Common families: ML (randomForest, glmnet, xgboost/gbm,
ranger, caret, pROC, Rtsne, e1071, dbscan), DSP
(signal, pracma, wavelets), causal (DoubleML, mlr3,
mlr3learners, ivreg, fixest), storage (RSQLite, duckdb).
library(rmorie)
manifest <- morie_read_outputs_manifest(project_root = "/path/to/project")
audit <- morie_audit_public_outputs(project_root = "/path/to/project",
manifest = manifest)
morie_summarize_output_audit(audit)library(rmorie)
synthetic_path <- morie_write_synthetic_data(
path = "data/private/synthetic_study_data.csv",
n = 8000,
seed = 2026,
overwrite = TRUE
)library(rmorie)
name_map <- morie_default_synthetic_name_map("generic")
name_map["cannabis_use"] <- "exposure_any"
name_map["bac"] <- "outcome_continuous"
dat <- morie_generate_synthetic_data(
n = 5000,
seed = 1,
name_map = name_map
)A first-class subsystem for the Ontario Special Investigations Unit director's-report corpus. The fetcher handles both English and French templates from 2005 onward; the parser is hand-rolled C++ for correctness under SIU's heterogeneous markup.
library(rmorie)
# Use the shipped language-aware DRID manifest; English-only,
# cache pages so re-runs are fast.
df <- morie_fetch_siu(
lang = "en", # skip French drids automatically
cache_html = TRUE, # persist every fetched page locally
rate_limit = 4 # requests per second (polite default)
)
# 2,218 unique cases x 64 columns; 100% format-clean on the
# shipping corpus per morie_siu_sanity_check().
nrow(df)# Inspect parser row + raw HTML + cleaned text side-by-side.
morie_siu_audit_case("16-OFI-019")
# Per-field "does the HTML actually support this value?" check.
morie_siu_anomaly_check("16-OFI-019")
# Diff parser output against an external table.
morie_siu_compare(
case_number = "16-OFI-019",
external = my_other_table,
field_map = c(officer_count = "n_officers")
)# Default: local Ollama with gemma3:4b. No API key required.
morie_siu_llm_extract("16-OFI-019")
# Failover chain: try local first, fall back to Gemini only on error.
morie_siu_llm_extract("16-OFI-019", model = c("ollama", "gemini"))
# French to English translation via translategemma.
morie_siu_translate(text = "L'enquete a ete close...", target_lang = "en")Supported providers: ollama (default), gemini, claude, vertex.
Environment knobs: OLLAMA_HOST (defaults to http://localhost:11434),
OLLAMA_MODEL (defaults to gemma3:4b), OLLAMA_KEEP_ALIVE (30m).
sane <- morie_siu_sanity_check(df)
sum(!sane$ok) # rows with format issues (regex / ISO date / Yes-No / chrome leak)# How accurate is each column across a sample of cases?
morie_siu_audit_columns(case_numbers = sample(df$case_number, 50))The parser learns. Ship-time corrections live in
inst/extdata/siu_canonical_overrides.csv.gz (47 hand-verified
corrections covering 10 spot-checked cases). Users can add their own:
morie_siu_record_correction(
case_number = "20-OFD-082",
field = "officer_count",
value = 3L
)Overrides are applied automatically at the end of morie_fetch_siu(),
per cell, by case number.
manifest <- morie_siu_index()
table(manifest$`_language`) # en=2531, fr=2212, unknown=0The R CMD check matrix covers six cells, all green on main:
| Platform | R version |
|---|---|
| macos-latest | release |
| windows-2025 | release |
| ubuntu-latest | release |
| ubuntu-latest | release + postgres-15 |
| ubuntu-latest | oldrel-1 |
| ubuntu-latest | devel |
Plus: pkgcheck, covr + Codecov upload, lintr, goodpractice, and
CodeQL.
If you use rmorie in your research, please cite the software:
Ruhela, V. S. (2026). rmorie: Multi-domain Open Research and Inferential Estimation in R. https://github.com/rootcoder007/rmorie
BibTeX (or run citation("rmorie") after installation for the entry
stamped with the exact installed version, sourced from inst/CITATION):
@Manual{ruhela_rmorie_2026,
title = {rmorie: Multi-domain Open Research and Inferential Estimation in R},
author = {Ruhela, Vansh Singh},
year = {2026},
url = {https://github.com/rootcoder007/rmorie}
}See CITATION.cff
for the machine-readable metadata GitHub's "Cite this repository" button uses.
R-MORIE is licensed under AGPL-3.0-or-later. See LICENSE for the
full text and LICENSING.md for the per-component breakdown.
rmorie's Bayesian regression (morie_bayes_lm) places zero-mean Normal
priors on the regression coefficients; the prior_sd argument is the
prior standard deviation (the scale of plausible coefficient values).
Larger prior_sd is weakly informative; smaller values pull estimates
toward zero (regularisation). Example:
d <- data.frame(x = rnorm(100)); d$y <- 1 + 2 * d$x + rnorm(100)
# weakly-informative prior (sd = 10) vs a tight regularising prior (sd = 0.5)
fit_weak <- morie_bayes_lm(y ~ x, d, prior_sd = 10)
fit_tight <- morie_bayes_lm(y ~ x, d, prior_sd = 0.5)See the bayesian-priors vignette for applied guidance.