cycleTrendR 0.3.0
Major new features
- Introduced universal time handling via the new
argument
dates_type.
- Full support for:
dates_type = "date" (daily data)dates_type = "posix" (sub-daily wearable/physiological
data)dates_type = "numeric" (neuroscience, simulations,
spike trains)
- Internal unified time index (
timenum) for consistent
modeling across formats.
Enhancements
- Automatic switching between STL, Lomb–Scargle, Fourier, LOESS, GAM,
and GAMM.
- Fourier harmonics now operate in time units of
timenum.
- Improved change-point detection compatible with all time
formats.
- Updated spectral analysis pipeline for irregular and numeric time
series.
- New vignette: cycleTrendR in practice.
Bug fixes
- Removed hard-coded assumptions about Date class.
- Fixed plotting issues related to time axis.
- Improved robustness of bootstrap confidence intervals.
Documentation
- Updated README with universal examples.
- Added new vignette demonstrating Date, POSIXct, and numeric
workflows.
cycleTrendR 0.2.0
- Major improvements to documentation, imports, and CRAN
compliance.
- Added full roxygen2 documentation for all parameters and return
values.
- Improved NAMESPACE management with explicit
@importFrom
directives.
- Enhanced vignette stability and reduced computational load in
examples.
- Achieved full CRAN compliance: 0 errors, 0 warnings, 0 notes.
cycleTrendR 0.1.0
- Initial release of cycleTrendR.
- Added the main function
adaptive_cycle_trend_analysis()
supporting:
- LOESS, GAM, and GAMM trend estimation
- Automatic Fourier harmonic selection (AICc/BIC)
- Lomb–Scargle periodogram for irregular sampling
- Bootstrap confidence intervals (IID and MBB)
- Change-point detection
- Rolling-origin forecasting
- Added publication-quality ggplot2 visualizations:
- Trend + CI
- Periodogram
- Residual ACF
- Diagnostics and summary tables
- Added a comprehensive vignette: cycleTrendR-overview.
- Added README with installation instructions and examples.