Update visualization/vintage_curve.py

visualization/vintage_curve.py

@@ -0,0 +1,299 @@
+# visualizations/vintage_curves.py
+
+import plotly.graph_objects as go
+import plotly.express as px
+from plotly.subplots import make_subplots
+import pandas as pd
+from metrics.metric_registry import METRIC_FUNCTIONS
+from analytics.performance_analysis import generate_metric_view
+
+
+def generate_delinquency_metric_chart(
+    df,
+    metric_name,
+    chart_type="line"
+):
+    """
+    Generate Plotly visualization for delinquency metrics across vintages.
+    
+    Args:
+        df: Master dataframe with booking vintage and metric columns
+        metric_name: Name of metric (e.g., "30+@3", "30+@6", "60+@6", "Yr1 NCL")
+        chart_type: "line" for vintage curves, "bar" for bar chart
+    
+    Returns:
+        Plotly figure object
+    """
+    
+    # Generate metric data without grouping (overall view)
+    result = generate_metric_view(
+        df=df,
+        metric_name=metric_name,
+        group_col=None
+    )
+    
+    # Sort by vintage
+    result = result.sort_values("booking_vintage")
+    
+    # Identify the rate column
+    rate_col = [
+        col for col in result.columns
+        if "rate" in col.lower()
+    ][0]
+    
+    if chart_type == "line":
+        
+        fig = go.Figure()
+        
+        fig.add_trace(
+            go.Scatter(
+                x=result["booking_vintage"],
+                y=result[rate_col],
+                mode="lines+markers",
+                name=metric_name,
+                line=dict(
+                    width=3,
+                    color="#1f77b4"
+                ),
+                marker=dict(
+                    size=8,
+                    symbol="circle"
+                ),
+                hovertemplate=(
+                    "<b>Vintage: %{x}</b><br>" +
+                    f"Rate: %{{y:.2f}}%<br>" +
+                    "<extra></extra>"
+                )
+            )
+        )
+        
+        fig.update_layout(
+            title=f"Vintage Curve: {metric_name}",
+            xaxis_title="Booking Vintage",
+            yaxis_title="Rate (%)",
+            hovermode="x unified",
+            plot_bgcolor="rgba(240,240,240,0.5)",
+            xaxis=dict(
+                showgrid=True,
+                gridwidth=1,
+                gridcolor="white"
+            ),
+            yaxis=dict(
+                showgrid=True,
+                gridwidth=1,
+                gridcolor="white"
+            ),
+            height=400,
+            template="plotly_white"
+        )
+        
+    else:  # bar chart
+        
+        fig = px.bar(
+            result,
+            x="booking_vintage",
+            y=rate_col,
+            title=f"Delinquency Rate by Vintage: {metric_name}",
+            labels={
+                "booking_vintage": "Booking Vintage",
+                rate_col: "Rate (%)"
+            },
+            color=rate_col,
+            color_continuous_scale="Reds",
+            text=rate_col
+        )
+        
+        fig.update_traces(
+            texttemplate="%{text:.2f}%",
+            textposition="outside"
+        )
+        
+        fig.update_layout(
+            hovermode="x unified",
+            height=400,
+            template="plotly_white"
+        )
+    
+    return fig
+
+
+def generate_multi_metric_comparison(
+    df,
+    metrics=None
+):
+    """
+    Generate comparison chart for multiple delinquency metrics across vintages.
+    
+    Args:
+        df: Master dataframe
+        metrics: List of metric names (default: ["30+@3", "30+@6", "60+@6", "Yr1 NCL"])
+    
+    Returns:
+        Plotly figure with subplots
+    """
+    
+    if metrics is None:
+        metrics = ["30+@3", "30+@6", "60+@6", "Yr1 NCL"]
+    
+    # Create subplots
+    fig = make_subplots(
+        rows=2,
+        cols=2,
+        subplot_titles=metrics,
+        specs=[
+            [{"secondary_y": False}, {"secondary_y": False}],
+            [{"secondary_y": False}, {"secondary_y": False}]
+        ]
+    )
+    
+    positions = [
+        (1, 1),
+        (1, 2),
+        (2, 1),
+        (2, 2)
+    ]
+    
+    for metric, (row, col) in zip(metrics, positions):
+        
+        result = generate_metric_view(
+            df=df,
+            metric_name=metric,
+            group_col=None
+        )
+        
+        result = result.sort_values("booking_vintage")
+        
+        rate_col = [
+            col_name for col_name in result.columns
+            if "rate" in col_name.lower()
+        ][0]
+        
+        fig.add_trace(
+            go.Scatter(
+                x=result["booking_vintage"],
+                y=result[rate_col],
+                mode="lines+markers",
+                name=metric,
+                line=dict(width=2),
+                marker=dict(size=6),
+                hovertemplate=(
+                    f"<b>{metric}</b><br>" +
+                    "Vintage: %{x}<br>" +
+                    "Rate: %{y:.2f}%<br>" +
+                    "<extra></extra>"
+                )
+            ),
+            row=row,
+            col=col
+        )
+        
+        fig.update_xaxes(
+            title_text="Vintage",
+            row=row,
+            col=col
+        )
+        
+        fig.update_yaxes(
+            title_text="Rate (%)",
+            row=row,
+            col=col
+        )
+    
+    fig.update_layout(
+        title_text="Delinquency Metrics Comparison Across Vintages",
+        height=800,
+        showlegend=False,
+        template="plotly_white",
+        hovermode="x unified"
+    )
+    
+    return fig
+
+
+def generate_segment_delinquency_curve(
+    df,
+    metric_name,
+    category
+):
+    """
+    Generate vintage curve for a specific delinquency metric segmented by category.
+    
+    Args:
+        df: Master dataframe
+        metric_name: Metric name
+        category: Segmentation category (e.g., "fico_band", "sourcing_channel")
+    
+    Returns:
+        Plotly figure with multiple traces (one per category value)
+    """
+    
+    result = generate_metric_view(
+        df=df,
+        metric_name=metric_name,
+        group_col=category
+    )
+    
+    result = result.sort_values(["booking_vintage", category])
+    
+    rate_col = [
+        col for col in result.columns
+        if "rate" in col.lower()
+    ][0]
+    
+    fig = go.Figure()
+    
+    # Get unique categories
+    categories = result[category].unique()
+    
+    # Color palette
+    colors = px.colors.qualitative.Set1
+    
+    for idx, cat in enumerate(sorted(categories)):
+        
+        cat_data = result[result[category] == cat]
+        color = colors[idx % len(colors)]
+        
+        fig.add_trace(
+            go.Scatter(
+                x=cat_data["booking_vintage"],
+                y=cat_data[rate_col],
+                mode="lines+markers",
+                name=str(cat),
+                line=dict(width=2, color=color),
+                marker=dict(size=6, color=color),
+                hovertemplate=(
+                    f"<b>{cat}</b><br>" +
+                    "Vintage: %{x}<br>" +
+                    "Rate: %{y:.2f}%<br>" +
+                    "<extra></extra>"
+                )
+            )
+        )
+    
+    fig.update_layout(
+        title=f"{metric_name} by {category.replace('_', ' ').title()}",
+        xaxis_title="Booking Vintage",
+        yaxis_title="Rate (%)",
+        hovermode="x unified",
+        plot_bgcolor="rgba(240,240,240,0.5)",
+        xaxis=dict(
+            showgrid=True,
+            gridwidth=1,
+            gridcolor="white"
+        ),
+        yaxis=dict(
+            showgrid=True,
+            gridwidth=1,
+            gridcolor="white"
+        ),
+        height=500,
+        template="plotly_white",
+        legend=dict(
+            x=1.05,
+            y=1,
+            xanchor="left",
+            yanchor="top"
+        )
+    )
+    
+    return fig