pwb-toolbox 0.1.9__py3-none-any.whl → 0.1.11__py3-none-any.whl

pwb_toolbox/backtest/__init__.py

@@ -1,3 +1,5 @@
 from .base_strategy import BaseStrategy
 from .engine import run_strategy
 from .ib_connector import IBConnector, run_ib_strategy
+from .portfolio import MonthlyEqualWeightPortfolio
+from .universe import get_most_liquid_symbols, get_least_volatile_symbols

pwb_toolbox/backtest/engine.py

@@ -3,7 +3,16 @@ import pandas as pd
 import pwb_toolbox.datasets as pwb_ds
 
 
-def run_strategy(signal, signal_kwargs, portfolio, symbols, start_date, leverage, cash):
+def run_strategy(
+    signal,
+    signal_kwargs,
+    portfolio,
+    portfolio_kwargs,
+    symbols,
+    start_date,
+    leverage,
+    cash,
+):
     """Run a tactical asset allocation strategy with Backtrader."""
     # Load the data from https://paperswithbacktest.com/datasets
     pivot_df = pwb_ds.get_pricing(
@@ -18,7 +27,7 @@ def run_strategy(signal, signal_kwargs, portfolio, symbols, start_date, leverage
     pivot_df.ffill(inplace=True)  # forward-fill holidays
     pivot_df.bfill(inplace=True)  # back-fill leading IPO gaps
     cerebro = bt.Cerebro()
-    for symbol in symbols:
+    for symbol in pivot_df.columns.levels[0]:
         data = bt.feeds.PandasData(dataname=pivot_df[symbol].copy())
         cerebro.adddata(data, name=symbol)
     cerebro.addstrategy(
@@ -27,6 +36,7 @@ def run_strategy(signal, signal_kwargs, portfolio, symbols, start_date, leverage
         leverage=0.9,
         signal_cls=signal,
         signal_kwargs=signal_kwargs,
+        **portfolio_kwargs,
     )
     cerebro.broker.set_cash(cash)
     strategy = cerebro.run()[0]

pwb_toolbox/backtest/portfolio.py

@@ -0,0 +1,30 @@
+from .base_strategy import BaseStrategy
+
+
+class MonthlyEqualWeightPortfolio(BaseStrategy):
+    params = (
+        ("leverage", 0.9),
+        ("signal_cls", None),
+        ("signal_kwargs", {}),
+    )
+
+    def __init__(self):
+        super().__init__()
+        self.sig = {
+            d._name: self.p.signal_cls(d, **self.p.signal_kwargs) for d in self.datas
+        }
+        self.last_month = -1
+
+    def next(self):
+        """Rebalance portfolio at the start of each month."""
+        super().next()
+        today = self.datas[0].datetime.date(0)
+        if today.month == self.last_month:
+            return
+        self.last_month = today.month
+        longs = [
+            d for d in self.datas if self.is_tradable(d) and self.sig[d._name][0] == 1
+        ]
+        wt = (self.p.leverage / len(longs)) if longs else 0.0
+        for d in self.datas:
+            self.order_target_percent(d, target=wt if d in longs else 0)

pwb_toolbox/backtest/universe.py

@@ -0,0 +1,37 @@
+from typing import List
+import pandas as pd
+import pwb_toolbox.datasets as pwb_ds
+
+
+def get_most_liquid_symbols(n: int = 1_200) -> List[str]:
+    """Return the `n` most liquid stock symbols (volume × price on last bar)."""
+    df = pwb_ds.load_dataset("Stocks-Daily-Price", adjust=True, extend=True)
+    last_date = df["date"].max()
+    today = df[df["date"] == last_date].copy()
+    today["liquidity"] = today["volume"] * today["close"]
+    liquid = today.sort_values("liquidity", ascending=False)
+    return liquid["symbol"].tolist()[:n]
+
+
+def get_least_volatile_symbols(symbols=["sp500"], start="1990-01-01") -> List[str]:
+    pivot = pwb_ds.get_pricing(
+        symbol_list=symbols,
+        fields=["open", "high", "low", "close"],
+        start_date=start,
+        extend=True,
+    )
+    td = pd.bdate_range(pivot.index.min(), pivot.index.max())
+    pivot = pivot.reindex(td).ffill().bfill()
+    symbols = []
+    for sym in pivot.columns.levels[0]:
+        df = (
+            pivot[sym]
+            .copy()
+            .reset_index()
+            .rename(columns={"index": "date"})
+            .set_index("date")
+        )
+        if df.close.pct_change().abs().max() > 3:  # same volatility filter
+            continue
+        symbols.append(sym)
+    return symbols

pwb_toolbox/datasets/__init__.py

@@ -921,6 +921,9 @@ def get_pricing(
         fields = ["close"]
     if isinstance(symbol_list, str):
         symbol_list = [symbol_list]
+    if isinstance(symbol_list, list) and "sp500" in symbol_list:
+        symbol_list.remove("sp500")
+        symbol_list += SP500_SYMBOLS
 
     fields = [f.lower() for f in fields]
     bad = [f for f in fields if f not in ALLOWED_FIELDS]
@@ -928,25 +931,32 @@ def get_pricing(
         raise ValueError(f"Invalid field(s): {bad}. Allowed: {sorted(ALLOWED_FIELDS)}")
 
     # --------------------------------------------------------------- download
-    DATASETS = [
-        ("Stocks-Daily-Price", extend),
-        ("ETFs-Daily-Price", extend),
-        ("Cryptocurrencies-Daily-Price", extend),
-        ("Bonds-Daily-Price", extend),
-        ("Commodities-Daily-Price", extend),
-        ("Forex-Daily-Price", extend),
-        ("Indices-Daily-Price", False),  # indices generally have no proxy data
-    ]
-    remaining = set(symbol_list)  # symbols still to fetch
+    universe = ds.load_dataset(
+        "paperswithbacktest/Universe-Daily-Price",
+        token=_get_hf_token(),
+    )
+    mapping = universe["train"].to_pandas()
+    mapping = mapping.set_index("symbol")["repo_id"].to_dict()
+
+    grouped = defaultdict(list)
+    remaining = []
+    for sym in symbol_list:
+        repo_id = mapping.get(sym)
+        repo_id = repo_id.split("/")[1] if isinstance(repo_id, str) else None
+        if repo_id:
+            grouped[repo_id].append(sym)
+        else:
+            print(f"Warning: No dataset found for symbol '{sym}'")
+            remaining.append(sym)
+
     frames = []
-    for dataset_name, ext_flag in DATASETS:
-        if not remaining:  # all symbols resolved → stop early
-            break
-        df_part = load_dataset(dataset_name, list(remaining), extend=ext_flag)
+    for repo_id, syms in grouped.items():
+        ext_flag = extend if repo_id != "Indices-Daily-Price" else False
+        df_part = load_dataset(repo_id, syms, extend=ext_flag)
         if not df_part.empty:
             frames.append(df_part)
-            remaining -= set(df_part["symbol"].unique())
-    df = pd.concat(frames, ignore_index=True)
+
+    df = pd.concat(frames, ignore_index=True) if frames else pd.DataFrame()
 
     df["date"] = pd.to_datetime(df["date"])
     df.set_index("date", inplace=True)

{pwb_toolbox-0.1.9.dist-info → pwb_toolbox-0.1.11.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pwb-toolbox
-Version: 0.1.9
+Version: 0.1.11
 Summary: A toolbox library for quant traders
 Home-page: https://github.com/paperswithbacktest/pwb-toolbox
 Author: Your Name

pwb_toolbox-0.1.11.dist-info/RECORD

@@ -0,0 +1,17 @@
+pwb_toolbox/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+pwb_toolbox/backtest/__init__.py,sha256=TZYVTvK-i0sjzVACvIfarfCZtoFtuPAZtia0rZ9RPm4,253
+pwb_toolbox/backtest/base_strategy.py,sha256=aKQ7XNxWkljaR7TQ-Tbs_uNWBDLdbmaLADcCsclBwuU,990
+pwb_toolbox/backtest/engine.py,sha256=LQXMw-CQrhOymVWwOA5wZXOBoo5L7ZNy9locJfDV7uU,1352
+pwb_toolbox/backtest/ib_connector.py,sha256=5T-pgT_MrDOxqdvXgT_hceIeewPs-rN3j4n-Wr-6JGU,2120
+pwb_toolbox/backtest/portfolio.py,sha256=e_hhy2695FOgDCFcHU8DhhSXc5We7sRC8O2NpDFdg48,930
+pwb_toolbox/backtest/universe.py,sha256=LzFBCOuKBVhgM4yMXOlTvPIrbevGkTS1itHIRxMHRRY,1272
+pwb_toolbox/datasets/__init__.py,sha256=gaPQDsQAdhEo-6_GuSPwrF6aHy9GTKENO-VJv8j9VU8,22527
+pwb_toolbox/performance/__init__.py,sha256=ds47RiOSL3iIwRE0S8dnGINcVPlZw_I9D21ueTSVP-I,2925
+pwb_toolbox/performance/metrics.py,sha256=szY8m45dZeJHciF4NxPxXlDyc78_5cLyIweRQJ_8lCE,15255
+pwb_toolbox/performance/plots.py,sha256=R6OV-SxJaJnBuJGh8XmsF58a7ERwn2Irf4zEqzGMRz4,12886
+pwb_toolbox/performance/trade_stats.py,sha256=I-iboKMwVLij6pc2r-KfNDnyF3LZV_LzzpgjIcJtgFw,4940
+pwb_toolbox-0.1.11.dist-info/licenses/LICENSE.txt,sha256=_Wjz7o7St3iVSPBRzE0keS8XSqSJ03A3NZ6cMlTaSK8,1079
+pwb_toolbox-0.1.11.dist-info/METADATA,sha256=67N8fbKy9je3OtCmdwL2NfkrQZ3s4JTPaEe5zmq87Fw,7131
+pwb_toolbox-0.1.11.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+pwb_toolbox-0.1.11.dist-info/top_level.txt,sha256=TZcXcF2AMkKkibZOuq6AYsHjajPgddHAGjQUT64OYGY,12
+pwb_toolbox-0.1.11.dist-info/RECORD,,

pwb_toolbox-0.1.9.dist-info/RECORD

@@ -1,15 +0,0 @@
-pwb_toolbox/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-pwb_toolbox/backtest/__init__.py,sha256=EDqoQ-P4ADV91F2oCgtSfFK1R_65BMVsFWDlk-zNHnM,128
-pwb_toolbox/backtest/base_strategy.py,sha256=aKQ7XNxWkljaR7TQ-Tbs_uNWBDLdbmaLADcCsclBwuU,990
-pwb_toolbox/backtest/engine.py,sha256=v5mO5HUhsQnxmBeSdkE0KwbYZnS6rB1TM_A3AYERa4s,1252
-pwb_toolbox/backtest/ib_connector.py,sha256=5T-pgT_MrDOxqdvXgT_hceIeewPs-rN3j4n-Wr-6JGU,2120
-pwb_toolbox/datasets/__init__.py,sha256=o2Q6nw8HmV_gTFfovhPJkoGdFsADBunFC4KqBl9Tpaw,22259
-pwb_toolbox/performance/__init__.py,sha256=ds47RiOSL3iIwRE0S8dnGINcVPlZw_I9D21ueTSVP-I,2925
-pwb_toolbox/performance/metrics.py,sha256=szY8m45dZeJHciF4NxPxXlDyc78_5cLyIweRQJ_8lCE,15255
-pwb_toolbox/performance/plots.py,sha256=R6OV-SxJaJnBuJGh8XmsF58a7ERwn2Irf4zEqzGMRz4,12886
-pwb_toolbox/performance/trade_stats.py,sha256=I-iboKMwVLij6pc2r-KfNDnyF3LZV_LzzpgjIcJtgFw,4940
-pwb_toolbox-0.1.9.dist-info/licenses/LICENSE.txt,sha256=_Wjz7o7St3iVSPBRzE0keS8XSqSJ03A3NZ6cMlTaSK8,1079
-pwb_toolbox-0.1.9.dist-info/METADATA,sha256=_Abqeq866im1F_F_x3iTEvV-Fg-QA0ri7d5xtAQ-NUk,7130
-pwb_toolbox-0.1.9.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-pwb_toolbox-0.1.9.dist-info/top_level.txt,sha256=TZcXcF2AMkKkibZOuq6AYsHjajPgddHAGjQUT64OYGY,12
-pwb_toolbox-0.1.9.dist-info/RECORD,,