Diagnostics
AnoFox Forecast diagnostics provides 8 functions across 3 categories: seasonality detection with 12 period detection algorithms, MSTL decomposition into trend/seasonal/residual components, and changepoint detection for identifying structural breaks. The ts_detect_periods_by function applies multiple detection methods simultaneously -- including autocorrelation, spectral analysis, and STL-based approaches -- to identify the dominant seasonal periods in each series. These diagnostics are designed to run before forecasting, ensuring you select the right model and parameters.
Discover patterns, anomalies, and structural breaks in your time series. AnoFox Forecast diagnostics cover seasonality detection, time series decomposition using MSTL, and changepoint detection for identifying regime shifts in your data.
Categories
Seasonality Detection
Detect and analyze seasonal patterns
detect_periods_byclassify_seasonalityTime Series Decomposition
Separate trend, seasonal, and residual
mstl_decomposition_bydetrend_byChangepoint Detection
Find structural breaks and regime shifts
detect_changepoints_byAll Functions
| Function | Category | Description |
|---|---|---|
anofox_fcst_ts_detect_periods_by | Seasonality | Multi-method period detection (12 algorithms) |
anofox_fcst_ts_classify_seasonality_by | Seasonality | Classify seasonality type per group |
anofox_fcst_ts_classify_seasonality | Seasonality | Classify seasonality (single series) |
anofox_fcst_ts_mstl_decomposition_by | Decomposition | MSTL decomposition (trend + seasonal + residual) |
anofox_fcst_ts_detrend_by | Decomposition | Remove trend using multiple methods |
anofox_fcst_ts_detect_peaks_by | Decomposition | Detect local maxima with prominence |
anofox_fcst_ts_analyze_peak_timing_by | Decomposition | Analyze peak timing variability |
anofox_fcst_ts_detect_changepoints_by | Changepoints | Multi-series changepoint detection |