openseries 1.5.7__py3-none-any.whl → 1.6.0__py3-none-any.whl

openseries/frame.py

@@ -6,34 +6,22 @@ from __future__ import annotations
 import datetime as dt
 from copy import deepcopy
 from functools import reduce
-from inspect import stack
 from logging import warning
-from pathlib import Path
-from typing import Callable, Optional, Union, cast
+from typing import Optional, Union, cast
 
 import statsmodels.api as sm  # type: ignore[import-untyped,unused-ignore]
 from numpy import (
-    append,
     array,
     cov,
     cumprod,
     divide,
-    dot,
-    float64,
-    inf,
     isinf,
-    linspace,
     log,
     nan,
     sqrt,
     square,
     std,
-    zeros,
 )
-from numpy import (
-    sum as npsum,
-)
-from numpy.typing import NDArray
 from pandas import (
     DataFrame,
     DatetimeIndex,
@@ -44,11 +32,7 @@ from pandas import (
     concat,
     merge,
 )
-from plotly.graph_objs import Figure  # type: ignore[import-untyped,unused-ignore]
-from plotly.io import to_html  # type: ignore[import-untyped,unused-ignore]
-from plotly.offline import plot  # type: ignore[import-untyped,unused-ignore]
-from pydantic import DirectoryPath, field_validator
-from scipy.optimize import minimize  # type: ignore[import-untyped,unused-ignore]
+from pydantic import field_validator
 
 # noinspection PyProtectedMember
 from statsmodels.regression.linear_model import (  # type: ignore[import-untyped,unused-ignore]
@@ -58,9 +42,7 @@ from typing_extensions import Self
 
 from openseries._common_model import _CommonModel
 from openseries.datefixer import do_resample_to_business_period_ends
-from openseries.load_plotly import load_plotly_dict
 from openseries.series import OpenTimeSeries
-from openseries.simulation import random_generator
 from openseries.types import (
     CountriesType,
     DaysInYearType,
@@ -68,18 +50,17 @@ from openseries.types import (
     LiteralCaptureRatio,
     LiteralFrameProps,
     LiteralHowMerge,
-    LiteralLinePlotMode,
     LiteralOlsFitCovType,
     LiteralOlsFitMethod,
     LiteralPandasReindexMethod,
-    LiteralPlotlyJSlib,
-    LiteralPlotlyOutput,
     LiteralPortfolioWeightings,
     LiteralTrunc,
     OpenFramePropertiesList,
     ValueType,
 )
 
+__all__ = ["OpenFrame"]
+
 
 # noinspection PyUnresolvedReferences
 class OpenFrame(_CommonModel):
@@ -478,12 +459,12 @@ class OpenFrame(_CommonModel):
             An OpenFrame object
 
         """
-        head = self.tsdf.loc[self.first_indices.max()].copy()
-        tail = self.tsdf.loc[self.last_indices.min()].copy()
+        head: Series[float] = self.tsdf.loc[self.first_indices.max()].copy()
+        tail: Series[float] = self.tsdf.loc[self.last_indices.min()].copy()
         dates = do_resample_to_business_period_ends(
             data=self.tsdf,
-            head=head,  # type: ignore[arg-type,unused-ignore]
-            tail=tail,  # type: ignore[arg-type,unused-ignore]
+            head=head,
+            tail=tail,
             freq=freq,
             countries=countries,
         )
@@ -1692,549 +1673,3 @@ class OpenFrame(_CommonModel):
         )
 
         return DataFrame(corrdf)
-
-
-def simulate_portfolios(
-    simframe: OpenFrame,
-    num_ports: int,
-    seed: int,
-) -> DataFrame:
-    """
-    Generate random weights for simulated portfolios.
-
-    Parameters
-    ----------
-    simframe: OpenFrame
-        Return data for portfolio constituents
-    num_ports: int
-        Number of possible portfolios to simulate
-    seed: int
-        The seed for the random process
-
-    Returns
-    -------
-    pandas.DataFrame
-        The resulting data
-
-    """
-    copi = simframe.from_deepcopy()
-
-    if any(
-        x == ValueType.PRICE for x in copi.tsdf.columns.get_level_values(1).to_numpy()
-    ):
-        copi.value_to_ret()
-        log_ret = copi.tsdf.copy()[1:]
-    else:
-        log_ret = copi.tsdf.copy()
-
-    log_ret.columns = log_ret.columns.droplevel(level=1)
-
-    randomizer = random_generator(seed=seed)
-
-    all_weights = zeros((num_ports, simframe.item_count))
-    ret_arr = zeros(num_ports)
-    vol_arr = zeros(num_ports)
-    sharpe_arr = zeros(num_ports)
-
-    for x in range(num_ports):
-        weights = array(randomizer.random(simframe.item_count))
-        weights = weights / npsum(weights)
-        all_weights[x, :] = weights
-
-        vol_arr[x] = sqrt(
-            dot(
-                weights.T,
-                dot(log_ret.cov() * simframe.periods_in_a_year, weights),
-            ),
-        )
-
-        ret_arr[x] = npsum(log_ret.mean() * weights * simframe.periods_in_a_year)
-
-        sharpe_arr[x] = ret_arr[x] / vol_arr[x]
-
-    # noinspection PyUnreachableCode
-    simdf = concat(
-        [
-            DataFrame({"stdev": vol_arr, "ret": ret_arr, "sharpe": sharpe_arr}),
-            DataFrame(all_weights, columns=simframe.columns_lvl_zero),
-        ],
-        axis="columns",
-    )
-    simdf = simdf.replace([inf, -inf], nan)
-    return simdf.dropna()
-
-
-def efficient_frontier(  # noqa: C901
-    eframe: OpenFrame,
-    num_ports: int = 5000,
-    seed: int = 71,
-    upperbounds: float = 1.0,
-    frontier_points: int = 200,
-    *,
-    tweak: bool = True,
-) -> tuple[DataFrame, DataFrame, NDArray[float64]]:
-    """
-    Identify an efficient frontier.
-
-    Parameters
-    ----------
-    eframe: OpenFrame
-        Portfolio data
-    num_ports: int, default: 5000
-        Number of possible portfolios to simulate
-    seed: int, default: 71
-        The seed for the random process
-    upperbounds: float, default: 1.0
-        The largest allowed allocation to a single asset
-    frontier_points: int, default: 200
-        number of points along frontier to optimize
-    tweak: bool, default: True
-        cutting the frontier to exclude multiple points with almost the same risk
-
-    Returns
-    -------
-    tuple[DataFrame, DataFrame, NDArray[float]]
-        The efficient frontier data, simulation data and optimal portfolio
-
-    """
-    if eframe.weights is None:
-        eframe.weights = [1.0 / eframe.item_count] * eframe.item_count
-
-    copi = eframe.from_deepcopy()
-
-    if any(
-        x == ValueType.PRICE for x in copi.tsdf.columns.get_level_values(1).to_numpy()
-    ):
-        copi.value_to_ret()
-        log_ret = copi.tsdf.copy()[1:]
-    else:
-        log_ret = copi.tsdf.copy()
-
-    log_ret.columns = log_ret.columns.droplevel(level=1)
-
-    simulated = simulate_portfolios(simframe=copi, num_ports=num_ports, seed=seed)
-
-    frontier_min = simulated.loc[simulated["stdev"].idxmin()]["ret"]
-    arithmetic_mean = log_ret.mean() * copi.periods_in_a_year
-    frontier_max = 0.0
-    if isinstance(arithmetic_mean, Series):
-        frontier_max = arithmetic_mean.max()
-
-    def _check_sum(weights: NDArray[float64]) -> float64:
-        return cast(float64, npsum(weights) - 1)
-
-    def _get_ret_vol_sr(
-        lg_ret: DataFrame,
-        weights: NDArray[float64],
-        per_in_yr: float,
-    ) -> NDArray[float64]:
-        ret = npsum(lg_ret.mean() * weights) * per_in_yr
-        volatility = sqrt(dot(weights.T, dot(lg_ret.cov() * per_in_yr, weights)))
-        sr = ret / volatility
-        return cast(NDArray[float64], array([ret, volatility, sr]))
-
-    def _diff_return(
-        lg_ret: DataFrame,
-        weights: NDArray[float64],
-        per_in_yr: float,
-        poss_return: float,
-    ) -> float64:
-        return cast(
-            float64,
-            _get_ret_vol_sr(lg_ret=lg_ret, weights=weights, per_in_yr=per_in_yr)[0]
-            - poss_return,
-        )
-
-    def _neg_sharpe(weights: NDArray[float64]) -> float64:
-        return cast(
-            float64,
-            _get_ret_vol_sr(
-                lg_ret=log_ret,
-                weights=weights,
-                per_in_yr=eframe.periods_in_a_year,
-            )[2]
-            * -1,
-        )
-
-    def _minimize_volatility(
-        weights: NDArray[float64],
-    ) -> float64:
-        return cast(
-            float64,
-            _get_ret_vol_sr(
-                lg_ret=log_ret,
-                weights=weights,
-                per_in_yr=eframe.periods_in_a_year,
-            )[1],
-        )
-
-    constraints = {"type": "eq", "fun": _check_sum}
-    bounds = tuple((0, upperbounds) for _ in range(eframe.item_count))
-    init_guess = array(eframe.weights)
-
-    opt_results = minimize(
-        fun=_neg_sharpe,
-        x0=init_guess,
-        method="SLSQP",
-        bounds=bounds,
-        constraints=constraints,
-    )
-
-    optimal = _get_ret_vol_sr(
-        lg_ret=log_ret,
-        weights=opt_results.x,
-        per_in_yr=eframe.periods_in_a_year,
-    )
-
-    frontier_y = linspace(start=frontier_min, stop=frontier_max, num=frontier_points)
-    frontier_x = []
-    frontier_weights = []
-
-    for possible_return in frontier_y:
-        cons = cast(
-            dict[str, Union[str, Callable[[float, NDArray[float64]], float64]]],
-            (
-                {"type": "eq", "fun": _check_sum},
-                {
-                    "type": "eq",
-                    "fun": lambda w, poss_return=possible_return: _diff_return(
-                        lg_ret=log_ret,
-                        weights=w,
-                        per_in_yr=eframe.periods_in_a_year,
-                        poss_return=poss_return,
-                    ),
-                },
-            ),
-        )
-
-        result = minimize(
-            fun=_minimize_volatility,
-            x0=init_guess,
-            method="SLSQP",
-            bounds=bounds,
-            constraints=cons,
-        )
-
-        frontier_x.append(result["fun"])
-        frontier_weights.append(result["x"])
-
-    # noinspection PyUnreachableCode
-    line_df = concat(
-        [
-            DataFrame(data=frontier_weights, columns=eframe.columns_lvl_zero),
-            DataFrame({"stdev": frontier_x, "ret": frontier_y}),
-        ],
-        axis="columns",
-    )
-    line_df["sharpe"] = line_df.ret / line_df.stdev
-
-    limit_small = 0.0001
-    line_df = line_df.mask(line_df.abs() < limit_small, 0.0)
-    line_df["text"] = line_df.apply(
-        lambda c: "<br><br>Weights:<br>"
-        + "<br>".join(
-            [f"{c[nm]:.1%}  {nm}" for nm in eframe.columns_lvl_zero],
-        ),
-        axis="columns",
-    )
-
-    if tweak:
-        limit_tweak = 0.001
-        line_df["stdev_diff"] = line_df.stdev.pct_change()
-        line_df = line_df.loc[line_df.stdev_diff.abs() > limit_tweak]
-        line_df = line_df.drop(columns="stdev_diff")
-
-    return line_df, simulated, append(optimal, opt_results.x)
-
-
-def constrain_optimized_portfolios(
-    data: OpenFrame,
-    serie: OpenTimeSeries,
-    portfolioname: str = "Current Portfolio",
-    simulations: int = 10000,
-    curve_points: int = 200,
-    upper_bound: float = 0.25,
-) -> tuple[OpenFrame, OpenTimeSeries, OpenFrame, OpenTimeSeries]:
-    """
-    Constrain optimized portfolios to those that improve on the current one.
-
-    Parameters
-    ----------
-    data: OpenFrame
-        Portfolio data
-    serie: OpenTimeSeries
-        A
-    portfolioname: str, default: "Current Portfolio"
-        Name of the portfolio
-    simulations: int, default: 10000
-        Number of possible portfolios to simulate
-    curve_points: int, default: 200
-        Number of optimal portfolios on the efficient frontier
-    upper_bound: float, default: 0.25
-        The largest allowed allocation to a single asset
-
-    Returns
-    -------
-    tuple[OpenFrame, OpenTimeSeries, OpenFrame, OpenTimeSeries]
-        The constrained optimal portfolio data
-
-    """
-    lr_frame = data.from_deepcopy()
-    mv_frame = data.from_deepcopy()
-
-    front_frame, sim_frame, optimal = efficient_frontier(
-        eframe=data,
-        num_ports=simulations,
-        frontier_points=curve_points,
-        upperbounds=upper_bound,
-    )
-
-    condition_least_ret = front_frame.ret > serie.arithmetic_ret
-    # noinspection PyArgumentList
-    least_ret_frame = front_frame[condition_least_ret].sort_values(by="stdev")
-    least_ret_port = least_ret_frame.iloc[0]
-    least_ret_port_name = f"Minimize vol & target return of {portfolioname}"
-    least_ret_weights = [least_ret_port[c] for c in lr_frame.columns_lvl_zero]
-    lr_frame.weights = least_ret_weights
-    resleast = OpenTimeSeries.from_df(lr_frame.make_portfolio(least_ret_port_name))
-
-    condition_most_vol = front_frame.stdev < serie.vol
-    # noinspection PyArgumentList
-    most_vol_frame = front_frame[condition_most_vol].sort_values(
-        by="ret",
-        ascending=False,
-    )
-    most_vol_port = most_vol_frame.iloc[0]
-    most_vol_port_name = f"Maximize return & target risk of {portfolioname}"
-    most_vol_weights = [most_vol_port[c] for c in mv_frame.columns_lvl_zero]
-    mv_frame.weights = most_vol_weights
-    resmost = OpenTimeSeries.from_df(mv_frame.make_portfolio(most_vol_port_name))
-
-    return lr_frame, resleast, mv_frame, resmost
-
-
-def prepare_plot_data(
-    assets: OpenFrame,
-    current: OpenTimeSeries,
-    optimized: NDArray[float64],
-) -> DataFrame:
-    """
-    Prepare date to be used as point_frame in the sharpeplot function.
-
-    Parameters
-    ----------
-    assets: OpenFrame
-        Portfolio data with individual assets and a weighted portfolio
-    current: OpenTimeSeries
-        The current or initial portfolio based on given weights
-    optimized: DataFrame
-        Data optimized with the efficient_frontier method
-
-    Returns
-    -------
-    DataFrame
-        The data prepared with mean returns, volatility and weights
-
-    """
-    txt = "<br><br>Weights:<br>" + "<br>".join(
-        [
-            f"{wgt:.1%}  {nm}"
-            for wgt, nm in zip(
-                cast(list[float], assets.weights),
-                assets.columns_lvl_zero,
-            )
-        ],
-    )
-
-    opt_text_list = [
-        f"{wgt:.1%}  {nm}" for wgt, nm in zip(optimized[3:], assets.columns_lvl_zero)
-    ]
-    opt_text = "<br><br>Weights:<br>" + "<br>".join(opt_text_list)
-    vol: Series[float] = assets.vol
-    plotframe = DataFrame(
-        data=[
-            assets.arithmetic_ret,
-            vol,
-            Series(
-                data=[""] * assets.item_count,
-                index=vol.index,
-            ),
-        ],
-        index=["ret", "stdev", "text"],
-    )
-    plotframe.columns = plotframe.columns.droplevel(level=1)
-    plotframe["Max Sharpe Portfolio"] = [optimized[0], optimized[1], opt_text]
-    plotframe[current.label] = [current.arithmetic_ret, current.vol, txt]
-
-    return plotframe
-
-
-def sharpeplot(  # noqa: C901
-    sim_frame: DataFrame = None,
-    line_frame: DataFrame = None,
-    point_frame: DataFrame = None,
-    point_frame_mode: LiteralLinePlotMode = "markers",
-    filename: Optional[str] = None,
-    directory: Optional[DirectoryPath] = None,
-    titletext: Optional[str] = None,
-    output_type: LiteralPlotlyOutput = "file",
-    include_plotlyjs: LiteralPlotlyJSlib = "cdn",
-    *,
-    title: bool = True,
-    add_logo: bool = True,
-    auto_open: bool = True,
-) -> tuple[Figure, str]:
-    """
-    Create scatter plot coloured by Sharpe Ratio.
-
-    Parameters
-    ----------
-    sim_frame: DataFrame, optional
-        Data from the simulate_portfolios method.
-    line_frame: DataFrame, optional
-        Data from the efficient_frontier method.
-    point_frame: DataFrame, optional
-        Data to highlight current and efficient portfolios.
-    point_frame_mode: LiteralLinePlotMode, default: markers
-        Which type of scatter to use.
-    filename: str, optional
-        Name of the Plotly html file
-    directory: DirectoryPath, optional
-        Directory where Plotly html file is saved
-    titletext: str, optional
-        Text for the plot title
-    output_type: LiteralPlotlyOutput, default: "file"
-        Determines output type
-    include_plotlyjs: LiteralPlotlyJSlib, default: "cdn"
-        Determines how the plotly.js library is included in the output
-    title: bool, default: True
-        Whether to add standard plot title
-    add_logo: bool, default: True
-        Whether to add Captor logo
-    auto_open: bool, default: True
-        Determines whether to open a browser window with the plot
-
-    Returns
-    -------
-    Figure
-        The scatter plot with simulated and optimized results
-
-    """
-    returns = []
-    risk = []
-
-    if directory:
-        dirpath = Path(directory).resolve()
-    elif Path.home().joinpath("Documents").exists():
-        dirpath = Path.home().joinpath("Documents")
-    else:
-        dirpath = Path(stack()[1].filename).parent
-
-    if not filename:
-        filename = "sharpeplot.html"
-    plotfile = dirpath.joinpath(filename)
-
-    fig, logo = load_plotly_dict()
-    figure = Figure(fig)
-
-    if sim_frame is not None:
-        returns.extend(list(sim_frame.loc[:, "ret"]))
-        risk.extend(list(sim_frame.loc[:, "stdev"]))
-        figure.add_scatter(
-            x=sim_frame.loc[:, "stdev"],
-            y=sim_frame.loc[:, "ret"],
-            hoverinfo="skip",
-            marker={
-                "size": 10,
-                "opacity": 0.5,
-                "color": sim_frame.loc[:, "sharpe"],
-                "colorscale": "Jet",
-                "reversescale": True,
-                "colorbar": {"thickness": 20, "title": "Ratio<br>ret / vol"},
-            },
-            mode="markers",
-            name="simulated portfolios",
-        )
-    if line_frame is not None:
-        returns.extend(list(line_frame.loc[:, "ret"]))
-        risk.extend(list(line_frame.loc[:, "stdev"]))
-        figure.add_scatter(
-            x=line_frame.loc[:, "stdev"],
-            y=line_frame.loc[:, "ret"],
-            text=line_frame.loc[:, "text"],
-            xhoverformat=".2%",
-            yhoverformat=".2%",
-            hovertemplate="Return %{y}<br>Vol %{x}%{text}",
-            hoverlabel_align="right",
-            line={"width": 2.5, "dash": "solid"},
-            mode="lines",
-            name="Efficient frontier",
-        )
-
-    colorway = cast(dict[str, list[str]], fig["layout"]).get("colorway")[
-        : len(point_frame.columns)
-    ]
-
-    if point_frame is not None:
-        for col, clr in zip(point_frame.columns, colorway):
-            returns.extend([point_frame.loc["ret", col]])
-            risk.extend([point_frame.loc["stdev", col]])
-            figure.add_scatter(
-                x=[point_frame.loc["stdev", col]],
-                y=[point_frame.loc["ret", col]],
-                xhoverformat=".2%",
-                yhoverformat=".2%",
-                hovertext=[point_frame.loc["text", col]],
-                hovertemplate="Return %{y}<br>Vol %{x}%{hovertext}",
-                hoverlabel_align="right",
-                marker={"size": 20, "color": clr},
-                mode=point_frame_mode,
-                name=col,
-                text=col,
-                textfont={"size": 14},
-                textposition="bottom center",
-            )
-
-    figure.update_layout(
-        xaxis={"tickformat": ".1%"},
-        xaxis_title="volatility",
-        yaxis={
-            "tickformat": ".1%",
-            "scaleanchor": "x",
-            "scaleratio": 1,
-        },
-        yaxis_title="annual return",
-        showlegend=False,
-    )
-    if title:
-        if titletext is None:
-            titletext = "<b>Risk and Return</b><br>"
-        figure.update_layout(title={"text": titletext, "font": {"size": 32}})
-
-    if add_logo:
-        figure.add_layout_image(logo)
-
-    if output_type == "file":
-        plot(
-            figure_or_data=figure,
-            filename=str(plotfile),
-            auto_open=auto_open,
-            auto_play=False,
-            link_text="",
-            include_plotlyjs=cast(bool, include_plotlyjs),
-            config=fig["config"],
-            output_type=output_type,
-        )
-        string_output = str(plotfile)
-    else:
-        div_id = filename.split(sep=".")[0]
-        string_output = to_html(
-            fig=figure,
-            config=fig["config"],
-            auto_play=False,
-            include_plotlyjs=cast(bool, include_plotlyjs),
-            full_html=False,
-            div_id=div_id,
-        )
-
-    return figure, string_output

openseries/load_plotly.py

@@ -11,6 +11,8 @@ from requests.exceptions import ConnectionError
 
 from openseries.types import CaptorLogoType, PlotlyLayoutType
 
+__all__ = ["load_plotly_dict"]
+
 
 def _check_remote_file_existence(url: str) -> bool:
     """