stochvolmodels 2.2.0__tar.gz → 2.3.0__tar.gz

{stochvolmodels-2.2.0 → stochvolmodels-2.3.0}/CHANGELOG.md

@@ -4,6 +4,38 @@ Entries start at 1.2.0. For earlier releases see the git log.
 
 ## [Unreleased]
 
+## [2.3.0] - 2026-08-24
+
+### Added
+- `stochvolmodels.estimation` now owns the Parkinson, Garman-Klass,
+  Rogers-Satchell, and close-to-close OHLC volatility estimators formerly maintained by `qis`.
+- `ForecastHorizon`, trading/calendar horizon presets, and the forward variance and volatility
+  target builders define point-in-time 1-day, 1-week, and 1-month forecast intervals without
+  including the information-time observation.
+- `estimate_ohlc_variances` and `build_volatility_features` provide a pooled OHLC variance panel
+  and append-only current, lagged, rolling-mean, LWMA, and downside-return forecast features.
+- `fit_volatility_forecaster` provides expanding-mean, persistence, fixed-decay EWMA, HAR, and
+  pooled OHLC level-NNLS fits with immutable diagnostics and non-negative predictions.
+- `walk_forward_volatility_forecast` adds expanding or rolling horizon-purged refits, immutable
+  per-origin and per-refit diagnostics, forecast losses, and common-sample benchmark comparisons.
+- `stochvolmodels.fitters.adapters.oca.option_chain_from_oca` is the reusable conversion from
+  one OCA point-in-time `SlicesChain` to SVM calibration inputs.
+- `plot_vix_1m_atm_vol.py` extracts and plots a same-session, constant-maturity VIX ATM
+  implied-volatility history from a partitioned OCA EOD cache without committing provider data.
+
+### Changed
+- `sample_option_chain_at_times` now delegates observation schedules and point-in-time selection
+  to OCA's `create_chain_timeseries`; `generate_vol_chain_np` remains a deprecated compatibility
+  wrapper around the fitter adapter.
+- `load_price_data` now delegates to the OCA chain object's linked underlying-data loader and is
+  retained only as a deprecated compatibility wrapper.
+- The `research` extra now requires `option-chain-analytics[cboe]>=5.2.0` and `qis>=5.11.0`,
+  including PyArrow support required by the local Parquet-backed option-data examples.
+
+### Removed
+- The legacy `dev` extra has been retired; contributor test dependencies now live in the
+  PEP 735 `test` dependency group.
+
 ## [2.2.0] - 2026-08-22
 
 ### Added

{stochvolmodels-2.2.0 → stochvolmodels-2.3.0}/MANIFEST.in

@@ -25,7 +25,6 @@ recursive-exclude examples *
 recursive-exclude papers *
 recursive-exclude docs *
 recursive-exclude resources *
-recursive-exclude volatility_book *
 recursive-exclude src/stochvolmodels/pde_solvers *
 
 global-exclude *.png

{stochvolmodels-2.2.0/src/stochvolmodels.egg-info → stochvolmodels-2.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: stochvolmodels
-Version: 2.2.0
+Version: 2.3.0
 Summary: Fourier-transform pricing, Monte Carlo validation, and calibration of European options under stochastic-volatility models in Python.
 Author-email: Artur Sepp <artursepp@gmail.com>
 Maintainer-email: Artur Sepp <artursepp@gmail.com>, Parviz Rakhmonov <ParvizRZ@gmail.com>
@@ -23,6 +23,7 @@ Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
 Classifier: Topic :: Office/Business :: Financial :: Investment
 Classifier: Topic :: Scientific/Engineering :: Mathematics
 Classifier: Topic :: Software Development :: Libraries :: Python Modules
@@ -37,8 +38,8 @@ Requires-Dist: pandas>=2.2.0
 Requires-Dist: matplotlib>=3.8.0
 Requires-Dist: seaborn>=0.13.0
 Provides-Extra: research
-Requires-Dist: option-chain-analytics>=5.0.0; extra == "research"
-Requires-Dist: qis>=3.5.7; extra == "research"
+Requires-Dist: option-chain-analytics[cboe]>=5.2.0; extra == "research"
+Requires-Dist: qis>=5.11.0; extra == "research"
 Provides-Extra: visualization
 Requires-Dist: plotly>=5.0.0; extra == "visualization"
 Provides-Extra: numerical
@@ -50,10 +51,6 @@ Requires-Dist: notebook>=6.5.0; extra == "jupyter"
 Requires-Dist: jupyterlab>=3.0.0; extra == "jupyter"
 Requires-Dist: ipykernel>=6.0.0; extra == "jupyter"
 Requires-Dist: ipywidgets>=8.0.0; extra == "jupyter"
-Provides-Extra: dev
-Requires-Dist: pytest>=7.0.0; extra == "dev"
-Requires-Dist: pytest-cov>=4.0.0; extra == "dev"
-Requires-Dist: pytest-regressions>=2.8.3; extra == "dev"
 Provides-Extra: docs
 Requires-Dist: sphinx<10.0,>=8.0; extra == "docs"
 Requires-Dist: furo>=2024.8.6; extra == "docs"
@@ -64,22 +61,22 @@ Dynamic: license-file
 
 # StochVolModels (`stochvolmodels`)
 
-`stochvolmodels` provides Fourier-transform pricing, Monte Carlo validation, and calibration of
-European options under stochastic-volatility models in Python.
-
-It is a focused research and practitioner library, not a general derivatives platform: the stable
-workflows cover European vanilla and related variance analytics under Heston and the
-Karasinski-Sepp log-normal stochastic-volatility model.
-
 [![PyPI](https://img.shields.io/pypi/v/stochvolmodels?style=flat-square)](https://pypi.org/project/stochvolmodels/)
 [![Python](https://img.shields.io/pypi/pyversions/stochvolmodels?style=flat-square)](https://pypi.org/project/stochvolmodels/)
 [![License](https://img.shields.io/github/license/ArturSepp/StochVolModels.svg?style=flat-square)](LICENSE.txt)
-[![CI](https://github.com/ArturSepp/StochVolModels/actions/workflows/ci.yml/badge.svg)](https://github.com/ArturSepp/StochVolModels/actions)
-[![Documentation](https://readthedocs.org/projects/stochvolmodels/badge/?version=latest)](https://stochvolmodels.readthedocs.io/en/latest/)
+[![CI](https://github.com/ArturSepp/StochVolModels/actions/workflows/ci.yml/badge.svg?branch=main)](https://github.com/ArturSepp/StochVolModels/actions/workflows/ci.yml)
+[![Docs](https://readthedocs.org/projects/stochvolmodels/badge/?version=latest)](https://stochvolmodels.readthedocs.io/en/latest/)
 [![Downloads](https://static.pepy.tech/badge/stochvolmodels)](https://pepy.tech/project/stochvolmodels)
 [![Monthly](https://static.pepy.tech/badge/stochvolmodels/month)](https://pepy.tech/project/stochvolmodels)
 [![Open LogSV quickstart in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/ArturSepp/StochVolModels/blob/main/examples/getting_started/quickstart_colab.ipynb)
 
+`stochvolmodels` provides Fourier-transform pricing, Monte Carlo validation, and calibration of
+European options under stochastic-volatility models in Python.
+
+It is a focused research and practitioner library, not a general derivatives platform: the stable
+workflows cover European vanilla and related variance analytics under Heston and the
+Karasinski-Sepp log-normal stochastic-volatility model.
+
 **Paper:** Sepp, A. and Rakhmonov, P. (2023), *Log-normal stochastic volatility model with quadratic drift*, [International Journal of Theoretical and Applied Finance, 26(8)](https://www.worldscientific.com/doi/10.1142/S0219024924500031). See [Citation](#citation) for the full BibTeX list.
 
 **Documentation:** [stochvolmodels.readthedocs.io](https://stochvolmodels.readthedocs.io/en/latest/) ·
@@ -145,7 +142,7 @@ python -m build
 python scripts/check_wheel_contents.py dist/*.whl
 ```
 
-The v2.2.0 quickstart prints two five-strike slices and deterministic reference values including
+The v2.3.0 quickstart prints two five-strike slices and deterministic reference values including
 `vanilla_price=0.197331` and `six_month_atm_price=0.275202`. The first call normally takes seconds
 because Numba compiles the numerical kernels. See the
 [full verification guide](https://stochvolmodels.readthedocs.io/en/latest/getting_started.html) and
@@ -166,7 +163,7 @@ because Numba compiles the numerical kernels. See the
 
 | Extra | Installs | Needed for |
 |---|---|---|
-| `research` | `qis >= 3.5.7`, `option-chain-analytics >= 5.0.0` | option-chain calibration and scripts in `papers/` |
+| `research` | `qis >= 5.11.0`, `option-chain-analytics[cboe] >= 5.2.0` | option-chain calibration, local Parquet data, and scripts in `papers/` |
 | `visualization` | `plotly >= 5.0.0` | interactive figures |
 | `numerical` | `scikit-learn >= 1.3.0`, `statsmodels >= 0.14.0` | statistical fits |
 | `jupyter` | `jupyter`, `notebook`, `jupyterlab`, `ipykernel`, `ipywidgets` | notebooks |
@@ -340,7 +337,7 @@ normalized cache under its `cboe_options/` subdirectory, and request only the ob
 needed by the experiment:
 
 ```bash
-pip install "stochvolmodels[research]" "option-chain-analytics[cboe]>=5.0.0"
+pip install "stochvolmodels[research]"
 ```
 
 ```python
@@ -676,7 +673,7 @@ If you use this package in your research, please cite the relevant papers:
   title={StochVolModels: Python Implementation of Stochastic Volatility Models},
   author={Sepp, Artur},
   year={2026},
-  version={2.2.0},
+  version={2.3.0},
   howpublished={\url{https://github.com/ArturSepp/StochVolModels}},
   note={Python package for pricing analytics and Monte Carlo simulations}
 }

{stochvolmodels-2.2.0 → stochvolmodels-2.3.0}/README.md

@@ -1,21 +1,21 @@
 # StochVolModels (`stochvolmodels`)
 
-`stochvolmodels` provides Fourier-transform pricing, Monte Carlo validation, and calibration of
-European options under stochastic-volatility models in Python.
-
-It is a focused research and practitioner library, not a general derivatives platform: the stable
-workflows cover European vanilla and related variance analytics under Heston and the
-Karasinski-Sepp log-normal stochastic-volatility model.
-
 [![PyPI](https://img.shields.io/pypi/v/stochvolmodels?style=flat-square)](https://pypi.org/project/stochvolmodels/)
 [![Python](https://img.shields.io/pypi/pyversions/stochvolmodels?style=flat-square)](https://pypi.org/project/stochvolmodels/)
 [![License](https://img.shields.io/github/license/ArturSepp/StochVolModels.svg?style=flat-square)](LICENSE.txt)
-[![CI](https://github.com/ArturSepp/StochVolModels/actions/workflows/ci.yml/badge.svg)](https://github.com/ArturSepp/StochVolModels/actions)
-[![Documentation](https://readthedocs.org/projects/stochvolmodels/badge/?version=latest)](https://stochvolmodels.readthedocs.io/en/latest/)
+[![CI](https://github.com/ArturSepp/StochVolModels/actions/workflows/ci.yml/badge.svg?branch=main)](https://github.com/ArturSepp/StochVolModels/actions/workflows/ci.yml)
+[![Docs](https://readthedocs.org/projects/stochvolmodels/badge/?version=latest)](https://stochvolmodels.readthedocs.io/en/latest/)
 [![Downloads](https://static.pepy.tech/badge/stochvolmodels)](https://pepy.tech/project/stochvolmodels)
 [![Monthly](https://static.pepy.tech/badge/stochvolmodels/month)](https://pepy.tech/project/stochvolmodels)
 [![Open LogSV quickstart in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/ArturSepp/StochVolModels/blob/main/examples/getting_started/quickstart_colab.ipynb)
 
+`stochvolmodels` provides Fourier-transform pricing, Monte Carlo validation, and calibration of
+European options under stochastic-volatility models in Python.
+
+It is a focused research and practitioner library, not a general derivatives platform: the stable
+workflows cover European vanilla and related variance analytics under Heston and the
+Karasinski-Sepp log-normal stochastic-volatility model.
+
 **Paper:** Sepp, A. and Rakhmonov, P. (2023), *Log-normal stochastic volatility model with quadratic drift*, [International Journal of Theoretical and Applied Finance, 26(8)](https://www.worldscientific.com/doi/10.1142/S0219024924500031). See [Citation](#citation) for the full BibTeX list.
 
 **Documentation:** [stochvolmodels.readthedocs.io](https://stochvolmodels.readthedocs.io/en/latest/) ·
@@ -81,7 +81,7 @@ python -m build
 python scripts/check_wheel_contents.py dist/*.whl
 ```
 
-The v2.2.0 quickstart prints two five-strike slices and deterministic reference values including
+The v2.3.0 quickstart prints two five-strike slices and deterministic reference values including
 `vanilla_price=0.197331` and `six_month_atm_price=0.275202`. The first call normally takes seconds
 because Numba compiles the numerical kernels. See the
 [full verification guide](https://stochvolmodels.readthedocs.io/en/latest/getting_started.html) and
@@ -102,7 +102,7 @@ because Numba compiles the numerical kernels. See the
 
 | Extra | Installs | Needed for |
 |---|---|---|
-| `research` | `qis >= 3.5.7`, `option-chain-analytics >= 5.0.0` | option-chain calibration and scripts in `papers/` |
+| `research` | `qis >= 5.11.0`, `option-chain-analytics[cboe] >= 5.2.0` | option-chain calibration, local Parquet data, and scripts in `papers/` |
 | `visualization` | `plotly >= 5.0.0` | interactive figures |
 | `numerical` | `scikit-learn >= 1.3.0`, `statsmodels >= 0.14.0` | statistical fits |
 | `jupyter` | `jupyter`, `notebook`, `jupyterlab`, `ipykernel`, `ipywidgets` | notebooks |
@@ -276,7 +276,7 @@ normalized cache under its `cboe_options/` subdirectory, and request only the ob
 needed by the experiment:
 
 ```bash
-pip install "stochvolmodels[research]" "option-chain-analytics[cboe]>=5.0.0"
+pip install "stochvolmodels[research]"
 ```
 
 ```python
@@ -612,7 +612,7 @@ If you use this package in your research, please cite the relevant papers:
   title={StochVolModels: Python Implementation of Stochastic Volatility Models},
   author={Sepp, Artur},
   year={2026},
-  version={2.2.0},
+  version={2.3.0},
   howpublished={\url{https://github.com/ArturSepp/StochVolModels}},
   note={Python package for pricing analytics and Monte Carlo simulations}
 }

{stochvolmodels-2.2.0 → stochvolmodels-2.3.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "stochvolmodels"
-version = "2.2.0"
+version = "2.3.0"
 description = "Fourier-transform pricing, Monte Carlo validation, and calibration of European options under stochastic-volatility models in Python."
 readme = "README.md"
 license = "MIT"
@@ -40,6 +40,7 @@ classifiers = [
     "Programming Language :: Python :: 3.11",
     "Programming Language :: Python :: 3.12",
     "Programming Language :: Python :: 3.13",
+    "Programming Language :: Python :: 3.14",
     "Topic :: Office/Business :: Financial :: Investment",
     "Topic :: Scientific/Engineering :: Mathematics",
     "Topic :: Software Development :: Libraries :: Python Modules",
@@ -58,8 +59,8 @@ dependencies = [
 
 [project.optional-dependencies]
 research = [
-    "option-chain-analytics>=5.0.0",
-    "qis>=3.5.7",
+    "option-chain-analytics[cboe]>=5.2.0",
+    "qis>=5.11.0",
 ]
 
 visualization = [
@@ -79,12 +80,6 @@ jupyter = [
     "ipywidgets>=8.0.0",
 ]
 
-dev = [
-    "pytest>=7.0.0",
-    "pytest-cov>=4.0.0",
-    "pytest-regressions>=2.8.3",
-]
-
 docs = [
     "sphinx>=8.0,<10.0",
     "furo>=2024.8.6",
@@ -97,6 +92,11 @@ all = [
 
 [dependency-groups]
 # Contributor-only tools are exact so local and CI verdicts are reproducible.
+test = [
+    "pytest>=7.0.0",
+    "pytest-cov>=4.0.0",
+    "pytest-regressions>=2.8.3",
+]
 lint = [
     "ruff==0.16.2",
     "interrogate==1.7.0",

{stochvolmodels-2.2.0 → stochvolmodels-2.3.0}/src/stochvolmodels/data/fetch_option_chain.py

@@ -12,32 +12,31 @@ import warnings
 from pathlib import Path
 from typing import Dict, Literal, Optional, Tuple
 
-import numpy as np
 import pandas as pd
-from numba.typed import List
 
 try:
-    import qis as qis
     from option_chain_analytics import (
         OptionsDataDFs,
+        SlicesChain,
         create_chain_at_time,
+        create_chain_timeseries,
     )
     from option_chain_analytics.data.cboe import load_local_cboe_options_data
     from option_chain_analytics.data.tardis import (
         load_local_tardis_contract_ts_data,
         load_local_tardis_eod_options_data,
     )
-    from option_chain_analytics.option_chain import SliceColumn, SlicesChain
     from qis import TimePeriod
 except ImportError as error:
     raise ImportError(
         "stochvolmodels.data.fetch_option_chain needs qis and option-chain-analytics: "
-        'pip install "stochvolmodels[research]" "option-chain-analytics[cboe]>=5.0.0"'
+        'pip install "stochvolmodels[research]" "option-chain-analytics[cboe]>=5.2.0"'
     ) from error
 
 # stochvolmodels
 from stochvolmodels import local_path as lp
 from stochvolmodels.data.option_chain import OptionChain
+from stochvolmodels.fitters.adapters.oca import option_chain_from_oca
 
 
 def _resolve_tardis_hourly_path(
@@ -345,74 +344,19 @@ def generate_vol_chain_np(chain: SlicesChain,
                           delta_bounds: Tuple[Optional[float], Optional[float]] = (-0.1, 0.1),
                           is_filtered: bool = True
                           ) -> OptionChain:
-    """Convert an OCA slices chain into SVM calibration inputs.
-
-    Normalized OCA panels supply positive fitted discount factors. Historical
-    hourly Tardis panels predate that column and retain their original unit-
-    discount convention.
-    """
-
-    records = []
-    seen_expiries = set()
-    for label, day in days_map.items():
-        next_date = value_time + pd.DateOffset(days=day)
-        slice_date = chain.get_next_slice_after_date(mat_date=next_date)
-        if slice_date in seen_expiries:
-            continue
-        slice_t = chain.expiry_slices[slice_date]
-        df = slice_t.get_joint_slice(delta_bounds=delta_bounds, is_filtered=is_filtered)
-        if not df.empty:
-            if SliceColumn.DISCOUNT.value in df.columns:
-                discounts = pd.to_numeric(
-                    df[SliceColumn.DISCOUNT.value], errors='coerce'
-                ).to_numpy()
-                discounts = discounts[np.isfinite(discounts) & (discounts > 0.0)]
-                if discounts.size == 0:
-                    raise ValueError(f"missing positive discount factor for {slice_t.expiry_id}")
-                discfactor = float(np.median(discounts))
-            else:
-                discfactor = 1.0
-            records.append(
-                {
-                    'id': f"{label}: {slice_t.expiry_id}",
-                    'ttm': slice_t.get_ttm(),
-                    'forward': slice_t.get_future_price(),
-                    'discfactor': discfactor,
-                    'strikes': df.index.to_numpy(),
-                    'optiontypes': df[SliceColumn.OPTION_TYPE].to_numpy(dtype=str),
-                    'bid_ivs': df[SliceColumn.BID_IV].to_numpy(),
-                    'ask_ivs': df[SliceColumn.ASK_IV].to_numpy(),
-                    'bid_prices': df[SliceColumn.BID_PRICE].to_numpy(),
-                    'ask_prices': df[SliceColumn.ASK_PRICE].to_numpy(),
-                }
-            )
-            seen_expiries.add(slice_date)
-
-    if not records:
-        raise ValueError("no non-empty OCA maturity slices matched days_map")
-    records.sort(key=lambda record: record['ttm'])
-    strikes_ttms, optiontypes_ttms = List(), List()
-    bid_ivs, ask_ivs = List(), List()
-    bid_prices, ask_prices = List(), List()
-    for record in records:
-        strikes_ttms.append(record['strikes'])
-        optiontypes_ttms.append(record['optiontypes'])
-        bid_ivs.append(record['bid_ivs'])
-        ask_ivs.append(record['ask_ivs'])
-        bid_prices.append(record['bid_prices'])
-        ask_prices.append(record['ask_prices'])
-
-    return OptionChain(
-        ttms=np.array([record['ttm'] for record in records]),
-        forwards=np.array([record['forward'] for record in records]),
-        discfactors=np.array([record['discfactor'] for record in records]),
-        ids=np.array([record['id'] for record in records]),
-        strikes_ttms=strikes_ttms,
-        optiontypes_ttms=optiontypes_ttms,
-        bid_ivs=bid_ivs,
-        ask_ivs=ask_ivs,
-        bid_prices=bid_prices,
-        ask_prices=ask_prices,
+    """Compatibility wrapper for :func:`option_chain_from_oca`."""
+    warnings.warn(
+        'generate_vol_chain_np moved to '
+        'stochvolmodels.fitters.adapters.oca.option_chain_from_oca',
+        DeprecationWarning,
+        stacklevel=2,
+    )
+    return option_chain_from_oca(
+        chain=chain,
+        value_time=value_time,
+        days_map=days_map,
+        delta_bounds=delta_bounds,
+        is_filtered=is_filtered,
     )
 
 
@@ -430,16 +374,14 @@ def load_option_chain(options_data_dfs: OptionsDataDFs,
         value_time=value_time,
         time_selection='previous',
     )
-    if chain is not None:
-        option_chain = generate_vol_chain_np(chain=chain,
-                                             value_time=chain.value_time,
-                                             days_map=days_map,
-                                             delta_bounds=delta_bounds,
-                                             is_filtered=is_filtered)
-    else:
-        option_chain = None
-
-    return option_chain
+    if chain is None:
+        return None
+    return option_chain_from_oca(
+        chain=chain,
+        days_map=days_map,
+        delta_bounds=delta_bounds,
+        is_filtered=is_filtered,
+    )
 
 
 def sample_option_chain_at_times(options_data_dfs: OptionsDataDFs,
@@ -452,32 +394,45 @@ def sample_option_chain_at_times(options_data_dfs: OptionsDataDFs,
                                  ),
                                  hour_offset: int = 8
                                  ) -> Dict[pd.Timestamp, OptionChain]:
-    """
-    extract chains at a sequence of observation times, for time series calibration.
-    """
-    value_times = qis.generate_dates_schedule(time_period=time_period,
-                                              freq=freq,
-                                              hour_offset=hour_offset)
-    option_chains = {}
-    for value_time in value_times:
-        option_chains[value_time] = load_option_chain(options_data_dfs=options_data_dfs,
-                                                      value_time=value_time,
-                                                      days_map=days_map,
-                                                      delta_bounds=delta_bounds,
-                                                      is_filtered=True)
-    return option_chains
+    """Compatibility workflow using OCA-owned observation-frequency sampling."""
+    warnings.warn(
+        'frequency sampling is owned by OptionChainAnalytics.create_chain_timeseries; '
+        'map its results with stochvolmodels.fitters.adapters.oca.option_chain_from_oca',
+        DeprecationWarning,
+        stacklevel=2,
+    )
+    chains = create_chain_timeseries(
+        options_data=options_data_dfs,
+        time_period=time_period,
+        freq=freq,
+        hour_offset=hour_offset,
+        time_selection='previous',
+    )
+    return {
+        value_time: option_chain_from_oca(
+            chain=chain,
+            days_map=days_map,
+            delta_bounds=delta_bounds,
+            is_filtered=True,
+        )
+        for value_time, chain in chains.items()
+    }
 
 
 def load_price_data(options_data_dfs: OptionsDataDFs,
                     time_period: TimePeriod = None,
                     data: Literal['close', 'perp', 'funding_rate'] = 'close',
-                    freq: Optional[str] = 'D'  # to do
+                    freq: Optional[str] = 'D'
                     ) -> pd.Series:
-    # options_data_dfs can also come from ts_data_loader_wrapper for legacy local sources.
-    """load the underlying price series accompanying the options data."""
-    spot_price = options_data_dfs.get_spot_data()[data]
-    if freq is not None:
-        spot_price = spot_price.resample(freq).last()
-    if time_period is not None:
-        spot_price = time_period.locate(spot_price)
-    return spot_price
+    """Compatibility wrapper for OCA's chain-linked underlying-data loader."""
+    warnings.warn(
+        'load_price_data moved to OptionChainAnalytics chain data: '
+        'options_data_dfs.load_price_data(...)',
+        DeprecationWarning,
+        stacklevel=2,
+    )
+    return options_data_dfs.load_price_data(
+        time_period=time_period,
+        data=data,
+        freq=freq,
+    )

stochvolmodels-2.3.0/src/stochvolmodels/estimation/__init__.py

@@ -0,0 +1,71 @@
+"""Statistical estimators used with stochastic-volatility models."""
+
+from stochvolmodels.estimation.forecasting import (
+    CALENDAR_1D,
+    CALENDAR_1M,
+    CALENDAR_1W,
+    CALENDAR_HORIZONS,
+    TRADING_1D,
+    TRADING_1M,
+    TRADING_1W,
+    TRADING_HORIZONS,
+    FeatureSpace,
+    ForecastHorizon,
+    ForecastSpace,
+    VolatilityForecastModel,
+    VolForecastComparison,
+    VolForecastConfig,
+    VolForecastEvaluation,
+    VolForecastFit,
+    VolForecastPrediction,
+    VolForecastResult,
+    build_volatility_features,
+    compare_volatility_forecasts,
+    evaluate_volatility_forecast,
+    fit_volatility_forecaster,
+    predict_volatility_forecaster,
+    walk_forward_volatility_forecast,
+)
+from stochvolmodels.estimation.ohlc import (
+    OhlcEstimatorType,
+    estimate_hf_ohlc_vol,
+    estimate_ohlc_var,
+    estimate_ohlc_variances,
+)
+from stochvolmodels.estimation.realized_volatility import (
+    make_forward_variance_target,
+    make_forward_volatility_target,
+)
+
+__all__ = [
+    "CALENDAR_1D",
+    "CALENDAR_1M",
+    "CALENDAR_1W",
+    "CALENDAR_HORIZONS",
+    "FeatureSpace",
+    "ForecastSpace",
+    "ForecastHorizon",
+    "OhlcEstimatorType",
+    "TRADING_1D",
+    "TRADING_1M",
+    "TRADING_1W",
+    "TRADING_HORIZONS",
+    "VolForecastConfig",
+    "VolForecastComparison",
+    "VolForecastEvaluation",
+    "VolForecastFit",
+    "VolForecastPrediction",
+    "VolForecastResult",
+    "VolatilityForecastModel",
+    "build_volatility_features",
+    "compare_volatility_forecasts",
+    "estimate_hf_ohlc_vol",
+    "estimate_ohlc_var",
+    "estimate_ohlc_variances",
+    "evaluate_volatility_forecast",
+    "fit_volatility_forecaster",
+    "make_forward_variance_target",
+    "make_forward_volatility_target",
+    "predict_volatility_forecaster",
+    "walk_forward_volatility_forecast",
+]

stochvolmodels-2.3.0/src/stochvolmodels/estimation/forecasting/__init__.py

@@ -0,0 +1,63 @@
+"""Point-in-time feature construction and volatility forecasting contracts."""
+
+from stochvolmodels.estimation.forecasting.features import (
+    FeatureSpace,
+    build_volatility_features,
+)
+from stochvolmodels.estimation.forecasting.horizons import (
+    CALENDAR_1D,
+    CALENDAR_1M,
+    CALENDAR_1W,
+    CALENDAR_HORIZONS,
+    TRADING_1D,
+    TRADING_1M,
+    TRADING_1W,
+    TRADING_HORIZONS,
+    ForecastHorizon,
+)
+from stochvolmodels.estimation.forecasting.models import (
+    fit_volatility_forecaster,
+    predict_volatility_forecaster,
+)
+from stochvolmodels.estimation.forecasting.results import (
+    ForecastSpace,
+    VolatilityForecastModel,
+    VolForecastComparison,
+    VolForecastConfig,
+    VolForecastEvaluation,
+    VolForecastFit,
+    VolForecastPrediction,
+    VolForecastResult,
+    compare_volatility_forecasts,
+    evaluate_volatility_forecast,
+)
+from stochvolmodels.estimation.forecasting.walk_forward import (
+    walk_forward_volatility_forecast,
+)
+
+__all__ = [
+    "CALENDAR_1D",
+    "CALENDAR_1M",
+    "CALENDAR_1W",
+    "CALENDAR_HORIZONS",
+    "FeatureSpace",
+    "ForecastSpace",
+    "ForecastHorizon",
+    "TRADING_1D",
+    "TRADING_1M",
+    "TRADING_1W",
+    "TRADING_HORIZONS",
+    "VolForecastConfig",
+    "VolForecastComparison",
+    "VolForecastEvaluation",
+    "VolForecastFit",
+    "VolForecastPrediction",
+    "VolForecastResult",
+    "VolatilityForecastModel",
+    "build_volatility_features",
+    "fit_volatility_forecaster",
+    "predict_volatility_forecaster",
+    "compare_volatility_forecasts",
+    "evaluate_volatility_forecast",
+    "walk_forward_volatility_forecast",
+]