Add files plot_memory.py and task_plot_memory.py for plotting the ACF of the differenced close price with confidence intervals.

Author: Lenr4

src/lennart_epp/final/plot_memory.py

@@ -0,0 +1,62 @@
+import numpy as np
+import plotly.graph_objects as go
+
+
+def plot_acf(acf_data: dict, save_path_html: str, save_path_pdf: str, n_obs: int):
+    """Plot the manually computed ACF using Plotly with confidence bands.
+
+    Args:
+        acf_data (dict): Dictionary containing "acf" values and "lags".
+        save_path_html (str): Path to save the interactive HTML plot.
+        save_path_pdf (str): Path to save the static PDF plot.
+        n_obs (int): Number of observations in the time series (for confidence bands).
+
+    Returns:
+        None: Saves plots in the specified paths.
+    """
+    max_lags = 200
+    lags = acf_data["lags"][:max_lags]
+    acf_values = acf_data["acf"][:max_lags]
+
+    confidence_band = 1.96 / np.sqrt(n_obs)
+
+    fig = go.Figure()
+
+    fig.add_trace(
+        go.Bar(
+            x=lags, y=acf_values, marker={"color": "rgba(0, 0, 255, 0.8)"}, name="ACF"
+        )
+    )
+
+    fig.add_trace(
+        go.Scatter(
+            x=lags,
+            y=[confidence_band] * len(lags),
+            mode="lines",
+            line={"color": "red", "dash": "dash"},
+            name="95% Confidence Interval",
+        )
+    )
+
+    fig.add_trace(
+        go.Scatter(
+            x=lags,
+            y=[-confidence_band] * len(lags),
+            mode="lines",
+            line={"color": "red", "dash": "dash"},
+            showlegend=False,
+        )
+    )
+
+    fig.update_layout(
+        title="Autocorrelation Function (ACF) with Confidence Bands",
+        xaxis_title="Lag",
+        yaxis_title="ACF Value",
+        template="plotly_white",
+        yaxis={"range": [-0.15, 0.3]},
+        plot_bgcolor="whitesmoke",
+        xaxis={"gridcolor": "lightgray"},
+    )
+
+    fig.write_html(save_path_html)
+    fig.write_image(save_path_pdf)

src/lennart_epp/final/task_plot_memory.py

@@ -0,0 +1,35 @@
+import pandas as pd
+
+from lennart_epp.config import BLD, SRC
+from lennart_epp.final.plot_memory import plot_acf
+
+
+def task_plot_acf(
+    script=SRC / "final" / "plot_memory.py",
+    acf_data=BLD / "memory" / "acf.pkl",
+    data=BLD / "data" / "cleaned_apple_data.pkl",
+    produces=(
+        BLD / "plots" / "acf_plot.html",
+        BLD / "plots" / "acf_plot.pdf",
+    ),
+):
+    """Task to generate and save the ACF plot for the differenced close price.
+
+    Args:
+        script (Path): Path to the script that generates the ACF plot.
+        acf_data (Path): Path to the computed ACF values stored in memory.
+        data (Path): Path to the original cleaned Apple stock data.
+        produces (tuple[Path, Path]): Paths to the output HTML and PDF plots.
+
+    Returns:
+        None: Ensures the plots are successfully generated.
+    """
+    acf_results = pd.read_pickle(acf_data)
+
+    df = pd.read_pickle(data)
+    n_obs = df["close_price"].diff().dropna().shape[0]
+
+    plot_acf(acf_results, produces[0], produces[1], n_obs)
+
+    for file in produces:
+        assert file.exists(), f"Failed to produce {file}"