Changes in version 0.1.0 (2026-04-21)                  

Initial CRAN release.

Core algorithm

  - iarimax() — per-subject ARIMAX fitting via forecast::auto.arima()
    followed by random-effects meta-analysis via metafor::rma().
    Supports multi-predictor models, optional fixed_d for cross-subject
    comparability, and keep_models for retaining raw model objects.

Preprocessing

  - i_screener() — pre-pipeline data quality screening on raw data.
    Three criteria: minimum observations (min_n_subject), minimum
    within-person SD (min_sd), and maximum modal response proportion
    (max_mode_pct). Supports "filter", "flag", and "report" output
    modes.

  - pmstandardize() — within-person z-scoring (person-mean centering and
    person-SD scaling).

  - i_detrender() — within-person linear detrending via lm(col ~
    timevar). Per-column filtering with pre- and post-detrend variance
    guards.

Inference

  - i_pval() — per-subject p-values using the two-tailed t-distribution
    with ML-based degrees of freedom (n_valid - n_params).

  - sden_test() — Sign Divergence Test (SDT) and Equisyncratic Null Test
    (ENT) with automatic selection based on the pooled REMA p-value.

Loop detection

  - looping_machine() — directed loop detection across three variables.
    Fits three iarimax() legs, applies i_pval(), and computes
    Loop_positive_directed.

S3 methods

  - summary.iarimax_results() — subject counts, direction/significance
    counts, REMA estimates, and heterogeneity statistics.

  - plot.iarimax_results() — caterpillar plot with per-subject
    confidence intervals and REMA band overlay.

  - summary.sden_results() — test type, hypothesis, and binomial test
    results.