davis-stats 2.2__tar.gz → 2.2.2__tar.gz

{davis_stats-2.2 → davis_stats-2.2.2}/PKG-INFO

@@ -1,17 +1,16 @@
 Metadata-Version: 2.4
 Name: davis_stats
-Version: 2.2
+Version: 2.2.2
 Summary: davis_stats is a teaching-focused repository 
 Author: Justin G. Davis
 Author-email: 
 Keywords: statistics,data science,business analytics
-Requires-Python: >=3.7
+Requires-Python: >=3.11
 Requires-Dist: pandas>=3.0.6
 Requires-Dist: openpyxl>=3.0.0
 Requires-Dist: numpy>=1.20.0
 Requires-Dist: matplotlib>=3.11.2
 Requires-Dist: scipy>=1.6.0
-Requires-Dist: seaborn>=0.12.0
 Requires-Dist: statsmodels>=0.14.0
 Requires-Dist: ipympl>=0.8.0
 Requires-Dist: linearmodels>=7.0

{davis_stats-2.2 → davis_stats-2.2.2}/davis_stats/__init__.py

@@ -19,4 +19,5 @@ from .datasets import (
     ceo_comp, netflix_content, olympic_medals, 
     restaurants, world_cup_goals, just_games, 
     nba, fired_ceo, restate, new_ceo, sp1500_2025,
-    sp1500_panel, student_athletes, m)
+    sp1500_panel, student_athletes, m, sweet_things,
+    sweet_things_simple)

{davis_stats-2.2 → davis_stats-2.2.2}/davis_stats/datasets/__init__.py

@@ -19,6 +19,8 @@ class DataFrames:
     _sp1500_panel = None
     _student_athletes = None
     _m = None
+    _sweet_things = None
+    _sweet_things_simple = None
 
     @property
     def ceo_comp(self):
@@ -104,6 +106,18 @@ class DataFrames:
             self._m = pd.read_excel(DATA_DIR / 'm.xlsx')
         return self._m
 
+    @property
+    def sweet_things(self):
+        if self._sweet_things is None:
+            self._sweet_things = pd.read_excel(DATA_DIR / 'sweet_things.xlsx')
+        return self._sweet_things
+
+    @property
+    def sweet_things_simple(self):
+        if self._sweet_things_simple is None:
+            self._sweet_things_simple = pd.read_excel(DATA_DIR / 'sweet_things_simple.xlsx')
+        return self._sweet_things_simple
+
 _datasets = DataFrames()
 
 
@@ -148,3 +162,9 @@ def student_athletes():
 
 def m():
     return _datasets.m
+
+def sweet_things():
+    return _datasets.sweet_things
+
+def sweet_things_simple():
+    return _datasets.sweet_things_simple

{davis_stats-2.2 → davis_stats-2.2.2}/davis_stats/visualization/boxplot.py

@@ -10,6 +10,7 @@ def boxplot(
     title=None,
     trim_outliers=100,
     details=True,
+    decimals=1,
     dpi=160,
     save_dpi=300,
     figsize=(7, 4.5),
@@ -79,7 +80,7 @@ def boxplot(
     ax.ticklabel_format(style="plain", axis=value_axis)
     ax.tick_params(axis=value_axis, labelsize=11, length=5, width=1.1)
     getattr(ax, f"{value_axis}axis").set_major_formatter(
-        FuncFormatter(lambda v, _: f"{v:,.2f}")
+        FuncFormatter(lambda v, _: f"{v:,.{decimals}f}")
     )
 
     if details:

{davis_stats-2.2 → davis_stats-2.2.2}/davis_stats/visualization/histogram.py

@@ -12,6 +12,7 @@ def histogram(
     bins=30,
     trim_outliers=100,
     details=True,
+    decimals=1,
     dpi=200,
     save_dpi=300,
     figsize=(7, 4.5),
@@ -55,7 +56,7 @@ def histogram(
     ax.set_xlabel("")
     ax.tick_params(labelsize=10, length=5, width=1.1)
     ax.ticklabel_format(style="plain", axis="x")
-    ax.xaxis.set_major_formatter(FuncFormatter(lambda v, _: f"{v:,.2f}"))
+    ax.xaxis.set_major_formatter(FuncFormatter(lambda v, _: f"{v:,.{decimals}f}"))
     ax.yaxis.set_major_formatter(FuncFormatter(lambda v, _: f"{v:,.0f}"))
     ax.yaxis.set_major_locator(MaxNLocator(nbins=6))
 

davis_stats-2.2.2/davis_stats/visualization/residuals.py

@@ -0,0 +1,74 @@
+import matplotlib.pyplot as plt
+import numpy as np
+from matplotlib.ticker import FuncFormatter, MaxNLocator
+from ..reg_modeling.reg import reg
+
+
+def residuals(
+    df,
+    y,
+    x,
+    dummies=None,
+    logistic=False,
+    decimals=1,
+    dpi=200,
+    save_dpi=300,
+    figsize=(7, 4.5),
+    save_path=None,
+    facecolor="white",
+):
+    """
+    Plot residuals vs fitted values to check linearity.
+    """
+    results = reg(df, y, x, dummies=dummies, logistic=logistic, silent=True)
+    if results is None:
+        print("Error: Could not fit regression model")
+        return
+
+    fitted = np.asarray(results.fittedvalues, dtype=float)
+    resid = np.asarray(results.resid, dtype=float)
+
+    fig, ax = plt.subplots(figsize=figsize, dpi=dpi, facecolor=facecolor)
+    fig.patch.set_facecolor(facecolor)
+    ax.set_facecolor(facecolor)
+
+    ax.scatter(
+        fitted,
+        resid,
+        s=32,
+        c="#7eb6d9",
+        alpha=0.78,
+        edgecolors="#1f1f1f",
+        linewidths=0.45,
+        zorder=3,
+    )
+    ax.axhline(
+        0, color="#c0392b", linestyle="--", linewidth=1.8, zorder=4,
+    )
+
+    fmt = FuncFormatter(lambda v, _: f"{v:,.{decimals}f}")
+    ax.grid(True, linestyle="--", linewidth=0.8, alpha=0.45)
+    ax.set_axisbelow(True)
+    ax.spines["top"].set_visible(False)
+    ax.spines["right"].set_visible(False)
+    ax.spines["left"].set_linewidth(1.2)
+    ax.spines["bottom"].set_linewidth(1.2)
+    ax.tick_params(labelsize=10, length=5, width=1.1)
+    ax.ticklabel_format(style="plain")
+    ax.xaxis.set_major_formatter(fmt)
+    ax.yaxis.set_major_formatter(fmt)
+    ax.xaxis.set_major_locator(MaxNLocator(nbins=6))
+    ax.yaxis.set_major_locator(MaxNLocator(nbins=6))
+
+    ax.set_title("Residuals vs Fitted Values", fontsize=13, pad=10)
+    ax.set_xlabel("Fitted Values", fontsize=11)
+    ax.set_ylabel("Residuals", fontsize=11)
+
+    fig.tight_layout()
+    if save_path is not None:
+        fig.savefig(
+            save_path, dpi=save_dpi, facecolor=facecolor,
+            bbox_inches="tight", pad_inches=0.15,
+        )
+    plt.show(block=False)
+    return ax

davis_stats-2.2.2/davis_stats/visualization/scatter.py

@@ -0,0 +1,194 @@
+import matplotlib.pyplot as plt
+import numpy as np
+import pandas as pd
+from matplotlib.ticker import FuncFormatter, MaxNLocator
+from mpl_toolkits.mplot3d import Axes3D  # noqa: F401
+
+
+def scatter(
+    df,
+    y,
+    x,
+    fit_line=False,
+    decimals=1,
+    dpi=200,
+    save_dpi=300,
+    figsize=(7, 4.5),
+    save_path=None,
+    facecolor="white",
+):
+    """
+    Create a 2D or 3D scatter plot with optional fit line.
+
+    fit_line:
+        False        — points only
+        "linear"     — OLS of y on x  (or a plane in 3D)
+        "quadratic"  — OLS of y on x and x²  (2D only)
+    """
+    if isinstance(x, str):
+        x = [x]
+
+    if len(x) == 1:
+        ax = _scatter_2d(
+            df, y, x[0], fit_line=fit_line, decimals=decimals,
+            dpi=dpi, figsize=figsize, facecolor=facecolor,
+        )
+    else:
+        ax = _scatter_3d(
+            df, y, x[0], x[1], fit_line=fit_line, decimals=decimals,
+            dpi=dpi, figsize=figsize, facecolor=facecolor,
+        )
+
+    fig = ax.figure
+    fig.tight_layout()
+    if save_path is not None:
+        fig.savefig(
+            save_path, dpi=save_dpi, facecolor=facecolor,
+            bbox_inches="tight", pad_inches=0.15,
+        )
+    plt.show(block=False)
+    return ax
+
+
+def _fit_kind(fit_line):
+    if not fit_line:
+        return None
+    if fit_line == "linear":
+        return "linear"
+    if fit_line == "quadratic":
+        return "quadratic"
+    raise ValueError("fit_line must be False, 'linear', or 'quadratic'")
+
+
+def _style_2d(ax, decimals):
+    fmt = FuncFormatter(lambda v, _: f"{v:,.{decimals}f}")
+    ax.grid(True, linestyle="--", linewidth=0.8, alpha=0.45)
+    ax.set_axisbelow(True)
+    ax.spines["top"].set_visible(False)
+    ax.spines["right"].set_visible(False)
+    ax.spines["left"].set_linewidth(1.2)
+    ax.spines["bottom"].set_linewidth(1.2)
+    ax.tick_params(labelsize=10, length=5, width=1.1)
+    ax.ticklabel_format(style="plain")
+    ax.xaxis.set_major_formatter(fmt)
+    ax.yaxis.set_major_formatter(fmt)
+    ax.xaxis.set_major_locator(MaxNLocator(nbins=6))
+    ax.yaxis.set_major_locator(MaxNLocator(nbins=6))
+
+
+def _scatter_2d(df, y, x_var, fit_line, decimals, dpi, figsize, facecolor):
+    plot_df = df[[x_var, y]].dropna()
+    if plot_df.empty:
+        raise ValueError("No data to plot after dropping NaNs.")
+
+    xv = plot_df[x_var].to_numpy(dtype=float)
+    yv = plot_df[y].to_numpy(dtype=float)
+    corr = float(pd.Series(xv).corr(pd.Series(yv)))
+
+    fig, ax = plt.subplots(figsize=figsize, dpi=dpi, facecolor=facecolor)
+    fig.patch.set_facecolor(facecolor)
+    ax.set_facecolor(facecolor)
+
+    ax.scatter(
+        xv,
+        yv,
+        s=32,
+        c="#7eb6d9",
+        alpha=0.78,
+        edgecolors="#1f1f1f",
+        linewidths=0.45,
+        zorder=3,
+    )
+
+    kind = _fit_kind(fit_line)
+    if kind is not None and len(xv) >= (3 if kind == "quadratic" else 2):
+        degree = 2 if kind == "quadratic" else 1
+        coefs = np.polyfit(xv, yv, degree)
+        x_line = np.linspace(xv.min(), xv.max(), 200)
+        y_line = np.polyval(coefs, x_line)
+        ax.plot(
+            x_line,
+            y_line,
+            color="#c0392b",
+            linewidth=1.8,
+            zorder=4,
+        )
+
+    _style_2d(ax, decimals)
+    ax.set_title(
+        f"{y} and {x_var}\nCorrelation: {corr:.2f}",
+        fontsize=13, pad=10,
+    )
+    ax.set_xlabel(x_var, fontsize=11)
+    ax.set_ylabel(y, fontsize=11)
+    return ax
+
+
+def _scatter_3d(df, y, x_var, z_var, fit_line, decimals, dpi, figsize, facecolor):
+    kind = _fit_kind(fit_line)
+    if kind == "quadratic":
+        raise ValueError("fit_line='quadratic' is only available for 2D scatter plots.")
+
+    plot_df = df[[x_var, z_var, y]].dropna()
+    if plot_df.empty:
+        raise ValueError("No data to plot after dropping NaNs.")
+
+    fig = plt.figure(figsize=figsize, dpi=dpi, facecolor=facecolor)
+    fig.patch.set_facecolor(facecolor)
+    ax = fig.add_subplot(111, projection="3d", facecolor=facecolor)
+
+    ax.scatter(
+        plot_df[x_var],
+        plot_df[z_var],
+        plot_df[y],
+        s=32,
+        c="#7eb6d9",
+        alpha=0.78,
+        edgecolors="#1f1f1f",
+        linewidths=0.45,
+        depthshade=True,
+    )
+
+    if kind == "linear":
+        X_data = np.column_stack(
+            [plot_df[x_var], plot_df[z_var], np.ones(len(plot_df))]
+        )
+        y_data = plot_df[y].to_numpy(dtype=float)
+        coeffs, *_ = np.linalg.lstsq(X_data, y_data, rcond=None)
+        a, b, c = coeffs
+
+        x_surf = np.linspace(plot_df[x_var].min(), plot_df[x_var].max(), 12)
+        z_surf = np.linspace(plot_df[z_var].min(), plot_df[z_var].max(), 12)
+        X_mesh, Z_mesh = np.meshgrid(x_surf, z_surf)
+        Y_mesh = a * X_mesh + b * Z_mesh + c
+        ax.plot_surface(
+            X_mesh,
+            Z_mesh,
+            Y_mesh,
+            color="#c0392b",
+            alpha=0.28,
+            edgecolor="none",
+            shade=False,
+        )
+
+    fmt = FuncFormatter(lambda v, _: f"{v:,.{decimals}f}")
+    ax.xaxis.set_major_formatter(fmt)
+    ax.yaxis.set_major_formatter(fmt)
+    ax.zaxis.set_major_formatter(fmt)
+
+    ax.set_title(f"{y}, {x_var}, and {z_var}", fontsize=13, pad=10)
+    ax.set_xlabel(x_var, fontsize=11, labelpad=8)
+    ax.set_ylabel(z_var, fontsize=11, labelpad=8)
+    ax.set_zlabel(y, fontsize=11, labelpad=8)
+    ax.tick_params(labelsize=9, length=4, width=1.0)
+
+    ax.zaxis._axinfo["juggled"] = (1, 2, 0)
+    ax.invert_xaxis()
+    ax.invert_yaxis()
+    ax.view_init(elev=25, azim=135)
+
+    grid = (0.55, 0.55, 0.55, 0.35)
+    ax.xaxis._axinfo["grid"]["color"] = grid
+    ax.yaxis._axinfo["grid"]["color"] = grid
+    ax.zaxis._axinfo["grid"]["color"] = grid
+    return ax

{davis_stats-2.2 → davis_stats-2.2.2}/davis_stats.egg-info/PKG-INFO

@@ -1,17 +1,16 @@
 Metadata-Version: 2.4
 Name: davis_stats
-Version: 2.2
+Version: 2.2.2
 Summary: davis_stats is a teaching-focused repository 
 Author: Justin G. Davis
 Author-email: 
 Keywords: statistics,data science,business analytics
-Requires-Python: >=3.7
+Requires-Python: >=3.11
 Requires-Dist: pandas>=3.0.6
 Requires-Dist: openpyxl>=3.0.0
 Requires-Dist: numpy>=1.20.0
 Requires-Dist: matplotlib>=3.11.2
 Requires-Dist: scipy>=1.6.0
-Requires-Dist: seaborn>=0.12.0
 Requires-Dist: statsmodels>=0.14.0
 Requires-Dist: ipympl>=0.8.0
 Requires-Dist: linearmodels>=7.0

{davis_stats-2.2 → davis_stats-2.2.2}/davis_stats.egg-info/SOURCES.txt

@@ -19,6 +19,8 @@ davis_stats/datasets/restaurants.xlsx
 davis_stats/datasets/sp1500_2025.xlsx
 davis_stats/datasets/sp1500_panel.xlsx
 davis_stats/datasets/student_athletes.xlsx
+davis_stats/datasets/sweet_things.xlsx
+davis_stats/datasets/sweet_things_simple.xlsx
 davis_stats/datasets/world_cup_goals.xlsx
 davis_stats/practice/__init__.py
 davis_stats/practice/stats_practice.py

{davis_stats-2.2 → davis_stats-2.2.2}/davis_stats.egg-info/requires.txt

@@ -3,7 +3,6 @@ openpyxl>=3.0.0
 numpy>=1.20.0
 matplotlib>=3.11.2
 scipy>=1.6.0
-seaborn>=0.12.0
 statsmodels>=0.14.0
 ipympl>=0.8.0
 linearmodels>=7.0

{davis_stats-2.2 → davis_stats-2.2.2}/setup.py

@@ -2,7 +2,7 @@ from setuptools import setup, find_packages
 
 setup(
     name='davis_stats',
-    version='2.2',
+    version='2.2.2',
     packages=find_packages(),
     package_data={
         'davis_stats': ['datasets/*.xlsx']},
@@ -12,11 +12,10 @@ setup(
         'numpy>=1.20.0',
         'matplotlib>=3.11.2',
         'scipy>=1.6.0',
-        'seaborn>=0.12.0',
         'statsmodels>=0.14.0',
         'ipympl>=0.8.0',
         'linearmodels>=7.0'],
-    python_requires='>=3.7',
+    python_requires='>=3.11',
     author='Justin G. Davis',
     author_email='',
     description='''

davis_stats-2.2/davis_stats/visualization/residuals.py

@@ -1,25 +0,0 @@
-import matplotlib.pyplot as plt
-from ..reg_modeling.reg import reg
-
-def residuals(df, y, x, dummies=None, logistic=False, dpi=150, figsize=(6, 4)):
-    """
-    Plot residuals vs fitted values to check linearity.
-    """
-    results = reg(df, y, x, dummies=dummies, logistic=logistic, silent=True)
-    
-    if results is None:
-        print("Error: Could not fit regression model")
-        return
-    
-    residuals = results.resid
-    fitted = results.fittedvalues
-    
-    plt.figure(figsize=figsize, dpi=dpi)
-    plt.scatter(fitted, residuals, alpha=0.5, edgecolors='k', linewidth=0.5)
-    plt.axhline(y=0, color='r', linestyle='--', linewidth=2)
-    plt.xlabel('Fitted Values', fontsize=12)
-    plt.ylabel('Residuals', fontsize=12)
-    plt.title('Residuals vs Fitted Values', fontsize=14, fontweight='bold')
-    plt.grid(True, alpha=0.3)
-    plt.tight_layout()
-    plt.show()

davis_stats-2.2/davis_stats/visualization/scatter.py

@@ -1,128 +0,0 @@
-import seaborn as sns
-import matplotlib.pyplot as plt
-import numpy as np
-from mpl_toolkits.mplot3d import Axes3D
-
-def scatter(df, y, x, fit_line=False, dpi=150, figsize=(6, 4)):
-    """
-    Create a 2D or 3D scatter plot with optional fit line.
-    """
-    
-    # Convert x to list if string
-    if isinstance(x, str):
-        x = [x]
-    
-    # Determine plot type based on number of x variables
-    if len(x) == 1:
-        # 2D scatter plot
-        x_var = x[0]
-        
-        # Calculate correlation coefficient
-        corr = df[x_var].corr(df[y])
-        
-        # Set style
-        sns.set_style("whitegrid")
-        
-        # Create figure
-        fig, ax = plt.subplots(figsize=figsize, dpi=dpi)
-        
-        # Create scatter plot
-        if fit_line:
-            sns.regplot(data=df, 
-                       x=x_var, 
-                       y=y,
-                       scatter_kws={'alpha':0.5},
-                       line_kws={'color': 'red'},
-                       ci=None)
-        else:
-            sns.scatterplot(data=df,
-                           x=x_var,
-                           y=y,
-                           alpha=0.5)
-        
-        # Customize plot
-        plt.title(f'{y} and {x_var}\nCorrelation: {corr:.3f}', pad=15)
-        plt.xlabel(x_var)
-        plt.ylabel(y)
-        
-        # Adjust layout
-        plt.tight_layout()
-        
-    else:
-        # 3D scatter plot (use first 2 variables from x list)
-        x_var = x[0]
-        z_var = x[1]
-        
-        # Create 3D figure
-        fig = plt.figure(figsize=figsize, dpi=dpi)
-        ax = fig.add_subplot(111, projection='3d')
-        
-        # Remove NaN values for plotting
-        plot_df = df[[x_var, z_var, y]].dropna()
-        
-        # Create 3D scatter plot
-        scatter = ax.scatter(plot_df[x_var], plot_df[z_var], plot_df[y], 
-                            c=plot_df[y], cmap='RdYlBu_r', s=50,
-                            edgecolor='black', linewidth=0.5,
-                            alpha=0.8)
-        
-        # Add colorbar
-        cbar = plt.colorbar(scatter, ax=ax, pad=0.15, shrink=0.8)
-        cbar.set_label(y, rotation=270, labelpad=15)
-        
-        # Add best-fit plane if requested
-        if fit_line:
-            # Prepare data for plane fitting: y = a*x + b*z + c
-            X_data = np.column_stack([plot_df[x_var], plot_df[z_var], np.ones(len(plot_df))])
-            y_data = plot_df[y].values
-            
-            # Fit plane using least squares
-            coeffs, residuals, rank, s = np.linalg.lstsq(X_data, y_data, rcond=None)
-            a, b, c = coeffs
-            
-            # Create mesh grid for the regression plane
-            x_surf = np.linspace(plot_df[x_var].min(), plot_df[x_var].max(), 20)
-            z_surf = np.linspace(plot_df[z_var].min(), plot_df[z_var].max(), 20)
-            X_mesh, Z_mesh = np.meshgrid(x_surf, z_surf)
-            
-            # Calculate y values for the plane
-            Y_mesh = a * X_mesh + b * Z_mesh + c
-            
-            # Plot the regression plane
-            surf = ax.plot_surface(X_mesh, Z_mesh, Y_mesh, 
-                                   alpha=0.4, cmap='coolwarm',
-                                   edgecolor='black', linewidth=0.5,
-                                   rstride=1, cstride=1,
-                                   antialiased=True)
-        
-        # Set title
-        title = f'{y}, {x_var}, and {z_var}'
-        
-        # Set axis labels
-        ax.set_xlabel(x_var, labelpad=10)
-        ax.set_ylabel(z_var, labelpad=10)
-        ax.set_zlabel(y, labelpad=10)
-        
-        # Move z-axis to left side
-        ax.zaxis._axinfo['juggled'] = (1, 2, 0)
-        
-        # Flip axes
-        ax.invert_xaxis()
-        ax.invert_yaxis()
-        
-        # Adjust viewing angle
-        ax.view_init(elev=25, azim=135)
-        
-        # Grid styling
-        ax.xaxis._axinfo["grid"]['color'] = (0.5, 0.5, 0.5, 0.5)
-        ax.yaxis._axinfo["grid"]['color'] = (0.5, 0.5, 0.5, 0.5)
-        ax.zaxis._axinfo["grid"]['color'] = (0.5, 0.5, 0.5, 0.5)
-        
-        # Set title
-        ax.set_title(title, pad=15)
-        
-        # Adjust layout
-        plt.tight_layout()
-    
-    # Show plot
-    plt.show()