"""Forecast methods registry and metadata."""
from typing import Any, Dict, List
import importlib.util as _importlib_util
# Availability checks
try:
from statsmodels.tsa.holtwinters import SimpleExpSmoothing as _SES, ExponentialSmoothing as _ETS # type: ignore
_SM_ETS_AVAILABLE = True
except Exception:
_SM_ETS_AVAILABLE = False
try:
from statsmodels.tsa.statespace.sarimax import SARIMAX as _SARIMAX # type: ignore
_SM_SARIMAX_AVAILABLE = True
except Exception:
_SM_SARIMAX_AVAILABLE = False
_NF_AVAILABLE = _importlib_util.find_spec("neuralforecast") is not None
_SF_AVAILABLE = _importlib_util.find_spec("statsforecast") is not None
_MLF_AVAILABLE = _importlib_util.find_spec("mlforecast") is not None
_LGB_AVAILABLE = _importlib_util.find_spec("lightgbm") is not None
_CHRONOS_AVAILABLE = _importlib_util.find_spec("chronos") is not None
_TIMESFM_AVAILABLE = _importlib_util.find_spec("timesfm") is not None
_LAG_LLAMA_AVAILABLE = (_importlib_util.find_spec("lag_llama") is not None)
def get_forecast_methods_data() -> Dict[str, Any]:
"""Return metadata about available forecast methods and their requirements."""
methods: List[Dict[str, Any]] = []
def add(method: str, description: str, params: List[Dict[str, Any]],
requires: List[str], supports: Dict[str, bool]) -> None:
available = True
reqs = list(requires)
if method in ("ses", "holt", "holt_winters_add", "holt_winters_mul") and not _SM_ETS_AVAILABLE:
available = False
reqs.append("statsmodels")
if method in ("arima", "sarima") and not _SM_SARIMAX_AVAILABLE:
available = False
reqs.append("statsmodels")
if method in ("nhits", "nbeatsx", "tft", "patchtst") and not _NF_AVAILABLE:
available = False
reqs.append("neuralforecast[torch]")
if method.startswith("sf_") and not _SF_AVAILABLE:
available = False
reqs.append("statsforecast")
if method == "mlf_rf" and not _MLF_AVAILABLE:
available = False
reqs.append("mlforecast, scikit-learn")
if method == "mlf_lightgbm" and (not _MLF_AVAILABLE or not _LGB_AVAILABLE):
available = False
reqs.append("mlforecast, lightgbm")
if method == "chronos_bolt" and not _CHRONOS_AVAILABLE:
available = False
reqs.append("chronos-forecasting")
if method == "timesfm" and not _TIMESFM_AVAILABLE:
available = False
reqs.append("timesfm")
if method == "lag_llama" and not _LAG_LLAMA_AVAILABLE:
available = False
reqs.append("lag-llama, gluonts, torch")
if method == "ensemble":
available = False
reqs.append("not implemented")
methods.append({
"method": method,
"available": bool(available),
"requires": sorted(set(reqs)),
"description": description,
"params": params,
"supports": supports,
})
# Baselines
add("naive", "Repeat last value forward.", [], [],
{"price": True, "return": True, "ci": True})
add("seasonal_naive", "Repeat last seasonal value (period m).", [
{"name": "seasonality", "type": "int", "default": None,
"description": "Seasonal period. Auto by timeframe if omitted."},
], [], {"price": True, "return": True, "ci": True})
add("drift", "Line from first to last with constant slope.", [], [],
{"price": True, "return": True, "ci": True})
# Classical
add("theta", "Theta method (SES + trend).", [
{"name": "seasonality", "type": "int", "default": None,
"description": "Seasonal period for deseasonalization."},
], [], {"price": True, "return": True, "ci": True})
add("fourier_ols", "Fourier series regression.", [
{"name": "seasonality", "type": "int", "default": None},
{"name": "fourier_order", "type": "int", "default": 5},
], [], {"price": True, "return": True, "ci": True})
# ETS
add("ses", "Simple Exponential Smoothing.", [
{"name": "alpha", "type": "float", "default": None},
], ["statsmodels"], {"price": True, "return": True, "ci": True})
add("holt", "Holt linear trend.", [
{"name": "alpha", "type": "float", "default": None},
{"name": "beta", "type": "float", "default": None},
], ["statsmodels"], {"price": True, "return": True, "ci": True})
add("holt_winters_add", "Holt-Winters additive.", [
{"name": "seasonality", "type": "int", "default": None},
], ["statsmodels"], {"price": True, "return": True, "ci": True})
add("holt_winters_mul", "Holt-Winters multiplicative.", [
{"name": "seasonality", "type": "int", "default": None},
], ["statsmodels"], {"price": True, "return": True, "ci": True})
# ARIMA
add("arima", "Non-seasonal ARIMA.", [
{"name": "p", "type": "int", "default": 1},
{"name": "d", "type": "int", "default": 1},
{"name": "q", "type": "int", "default": 1},
], ["statsmodels"], {"price": True, "return": True, "ci": True})
add("sarima", "Seasonal ARIMA.", [
{"name": "p", "type": "int", "default": 1},
{"name": "d", "type": "int", "default": 1},
{"name": "q", "type": "int", "default": 1},
{"name": "P", "type": "int", "default": 1},
{"name": "D", "type": "int", "default": 1},
{"name": "Q", "type": "int", "default": 1},
{"name": "seasonality", "type": "int", "default": None},
], ["statsmodels"], {"price": True, "return": True, "ci": True})
# Monte Carlo
add("mc_gbm", "Geometric Brownian Motion Monte Carlo.", [
{"name": "n_sims", "type": "int", "default": 1000},
{"name": "seed", "type": "int", "default": None},
], [], {"price": True, "return": False, "ci": True, "distribution": True})
add("hmm_mc", "HMM-based regime Monte Carlo.", [
{"name": "n_states", "type": "int", "default": 2},
{"name": "n_sims", "type": "int", "default": 1000},
{"name": "seed", "type": "int", "default": None},
], [], {"price": True, "return": False, "ci": True, "distribution": True})
# StatsForecast
add("sf_autoarima", "StatsForecast AutoARIMA.", [
{"name": "seasonality", "type": "int", "default": None},
{"name": "stepwise", "type": "bool", "default": True},
], ["statsforecast"], {"price": True, "return": True, "ci": True})
add("sf_theta", "StatsForecast Theta.", [
{"name": "seasonality", "type": "int", "default": None},
], ["statsforecast"], {"price": True, "return": True, "ci": True})
add("sf_autoets", "StatsForecast AutoETS.", [
{"name": "seasonality", "type": "int", "default": None},
], ["statsforecast"], {"price": True, "return": True, "ci": True})
add("sf_seasonalnaive", "StatsForecast SeasonalNaive.", [
{"name": "seasonality", "type": "int", "default": None},
], ["statsforecast"], {"price": True, "return": True, "ci": True})
# MLForecast
add("mlf_rf", "MLForecast Random Forest.", [
{"name": "lags", "type": "list", "default": [1, 2, 3]},
{"name": "n_estimators", "type": "int", "default": 100},
], ["mlforecast", "scikit-learn"], {"price": True, "return": True, "ci": False})
add("mlf_lightgbm", "MLForecast LightGBM.", [
{"name": "lags", "type": "list", "default": [1, 2, 3]},
{"name": "n_estimators", "type": "int", "default": 100},
], ["mlforecast", "lightgbm"], {"price": True, "return": True, "ci": False})
# NeuralForecast
add("nhits", "NeuralForecast NHITS.", [
{"name": "max_epochs", "type": "int", "default": 50},
{"name": "input_size", "type": "int", "default": 128},
], ["neuralforecast[torch]"], {"price": True, "return": True, "ci": True})
add("nbeatsx", "NeuralForecast NBEATS-X.", [
{"name": "max_epochs", "type": "int", "default": 50},
{"name": "input_size", "type": "int", "default": 128},
], ["neuralforecast[torch]"], {"price": True, "return": True, "ci": True})
add("tft", "NeuralForecast TFT.", [
{"name": "max_epochs", "type": "int", "default": 50},
{"name": "input_size", "type": "int", "default": 128},
], ["neuralforecast[torch]"], {"price": True, "return": True, "ci": True})
add("patchtst", "NeuralForecast PatchTST.", [
{"name": "max_epochs", "type": "int", "default": 50},
{"name": "input_size", "type": "int", "default": 128},
], ["neuralforecast[torch]"], {"price": True, "return": True, "ci": True})
# Foundation models
add("chronos_bolt", "Chronos-Bolt foundation model.", [
{"name": "model_name", "type": "str", "default": "amazon/chronos-bolt-base"},
{"name": "context_length", "type": "int", "default": 512},
], ["chronos-forecasting"], {"price": True, "return": True, "ci": True})
add("timesfm", "Google TimesFM foundation model.", [
{"name": "context_length", "type": "int", "default": 512},
], ["timesfm"], {"price": True, "return": True, "ci": True})
add("lag_llama", "Lag-Llama foundation model.", [
{"name": "context_length", "type": "int", "default": 512},
], ["lag-llama", "gluonts", "torch"], {"price": True, "return": True, "ci": True})
return {"methods": methods}
