From 942dd6c3818e0fd1f9c7a168bae33f156d5d0448 Mon Sep 17 00:00:00 2001
From: Thomas Schmelzer <thomas.schmelzer@gmail.com>
Date: Thu, 26 Dec 2024 21:29:43 +0400
Subject: [PATCH] towards marimo

---
 book/docs/marimo/assign.py |  57 ++++++++++
 book/docs/marimo/demo.py   | 210 +++++++++++++++++++++++++++++++++++++
 2 files changed, 267 insertions(+)
 create mode 100644 book/docs/marimo/assign.py
 create mode 100644 book/docs/marimo/demo.py

diff --git a/book/docs/marimo/assign.py b/book/docs/marimo/assign.py
new file mode 100644
index 00000000..107bd7ad
--- /dev/null
+++ b/book/docs/marimo/assign.py
@@ -0,0 +1,57 @@
+import marimo
+
+__generated_with = "0.9.27"
+app = marimo.App()
+
+
+@app.cell
+def __(mo):
+    mo.md(
+        r"""
+        # Reset a large matrix
+        """
+    )
+    return
+
+
+@app.cell
+def __():
+    import numpy as np
+
+    return (np,)
+
+
+@app.cell
+def __(np):
+    a = np.random.randn(5000, 5000)
+    _b = np.random.randn(5000, 5000)
+    return (a,)
+
+
+@app.cell
+def __(a):
+    _b = 0 * a
+    return
+
+
+@app.cell
+def __(a, np):
+    _b = np.zeros_like(a)
+    return
+
+
+@app.cell
+def __(np):
+    _b = np.zeros((5000, 5000))
+    return
+
+
+@app.cell
+def __():
+    import marimo as mo
+
+    return (mo,)
+
+
+if __name__ == "__main__":
+    app.run()
diff --git a/book/docs/marimo/demo.py b/book/docs/marimo/demo.py
new file mode 100644
index 00000000..35854d6f
--- /dev/null
+++ b/book/docs/marimo/demo.py
@@ -0,0 +1,210 @@
+import marimo
+
+__generated_with = "0.9.27"
+app = marimo.App()
+
+
+@app.cell
+def __(mo):
+    mo.md(
+        r"""
+        # Long only 1/n portfolio
+        """
+    )
+    return
+
+
+@app.cell
+def __():
+    import pandas as pd
+
+    pd.options.plotting.backend = "plotly"
+
+    import yfinance as yf
+
+    from cvx.simulator.builder import builder
+    from cvx.simulator.grid import resample_index
+
+    return builder, pd, resample_index, yf
+
+
+@app.cell
+def __(yf):
+    data = yf.download(
+        tickers="SPY AAPL GOOG MSFT",  # list of tickers
+        period="10y",  # time period
+        interval="1d",  # trading interval
+        prepost=False,  # download pre/post market hours data?
+        repair=True,
+    )  # repair obvious price errors e.g. 100x?
+    return (data,)
+
+
+@app.cell
+def __(data):
+    prices = data["Adj Close"]
+    return (prices,)
+
+
+@app.cell
+def __():
+    capital = 1e6
+    return (capital,)
+
+
+@app.cell
+def __(builder, capital, prices, state, time):
+    b = builder(prices=prices, initial_cash=capital)
+    for _time, _state in b:
+        b[time[-1]] = 0.25 * state.nav / state.prices
+    return (b,)
+
+
+@app.cell
+def __(b):
+    portfolio = b.build()
+    portfolio.profit.cumsum().plot()
+    return (portfolio,)
+
+
+@app.cell
+def __(portfolio):
+    portfolio.nav.plot()
+    return
+
+
+@app.cell
+def __(mo):
+    mo.md(
+        r"""
+        ## Rebalancing
+        """
+    )
+    return
+
+
+@app.cell
+def __(mo):
+    mo.md(
+        r"""
+        Usually we would not execute on a daily basis but rather rebalance every week, month or quarter.
+        There are two approaches to deal with this problem in cvxsimulator.
+
+        * Resample the existing daily portfolio (helpful to see effect of your hesitated trading)
+        * Trade only on days that are within a predefined grid (most flexible if you have a rather irregular grid)
+        """
+    )
+    return
+
+
+@app.cell
+def __(mo):
+    mo.md(
+        r"""
+        ### Resample an existing portfolio
+        """
+    )
+    return
+
+
+@app.cell
+def __(portfolio):
+    portfolio_resampled = portfolio.resample(rule="M")
+    return (portfolio_resampled,)
+
+
+@app.cell
+def __(pd, portfolio, portfolio_resampled):
+    frame = pd.DataFrame(
+        {"original": portfolio.nav, "monthly": portfolio_resampled.nav}
+    )
+    frame
+    return (frame,)
+
+
+@app.cell
+def __(portfolio_resampled):
+    print(portfolio_resampled.stocks)
+    return
+
+
+@app.cell
+def __(frame):
+    # almost hard to see that difference between the original and resampled portfolio
+    frame.plot()
+    return
+
+
+@app.cell
+def __(portfolio_resampled):
+    # number of shares traded
+    portfolio_resampled.trades_stocks.iloc[1:].plot()
+    return
+
+
+@app.cell
+def __(mo):
+    mo.md(
+        r"""
+        ### Trade only days in predefined grid
+        """
+    )
+    return
+
+
+@app.cell
+def __(builder, capital, prices, resample_index, state, time):
+    b_1 = builder(prices=prices, initial_cash=capital)
+    grid = resample_index(prices.index, rule="M")
+    for _time, _state in b_1:
+        if _time[-1] in grid:
+            b_1[time[-1]] = 0.25 * state.nav / state.prices
+        else:
+            b_1[time[-1]] = b_1[time[-2]]
+    portfolio_1 = b_1.build()
+    return b_1, grid, portfolio_1
+
+
+@app.cell
+def __(portfolio_1):
+    portfolio_1.nav.plot()
+    return
+
+
+@app.cell
+def __(portfolio_1):
+    portfolio_1.turnover.iloc[1:].plot()
+    return
+
+
+@app.cell
+def __(mo):
+    mo.md(
+        r"""
+        ### Why not resampling the prices?
+        """
+    )
+    return
+
+
+@app.cell
+def __(mo):
+    mo.md(
+        r"""
+        I don't believe in bringing the prices to a monthly grid. This would render it hard to construct signals
+        given the sparse grid. We stay on a daily grid and trade once a month.
+
+        """
+    )
+    return
+
+
+@app.cell
+def __():
+    import marimo as mo
+
+    return (mo,)
+
+
+if __name__ == "__main__":
+    app.run()