Distributional charts

policyengine/charts/distributional_impact/by_income_decile/average.py

@@ -0,0 +1,114 @@
+import plotly.graph_objects as go
+from policyengine_core.charts.formatting import *
+
+class ByIncomeDecileAverageChart:
+    def __init__(self, country: str, data=None):
+        if data is None:
+            raise ValueError("Data must be provided")
+
+        # Store values as they are (no percentage conversion)
+        for i in range(1, 12):
+            setattr(self, f"decile_{i}", data['average'][i])
+
+        self.country = country
+
+    def _get_color(self, value):
+        if value is None or value == 0 or value < 0:
+            return GRAY
+        return BLUE
+
+    def _get_change_direction(self, value):
+        if value > 0:
+            return "increase"
+        elif value < 0:
+            return "decrease"
+        else:
+            return "no change"
+
+    def _get_currency_symbol(self):
+        if self.country.lower() == "us":
+            return "$"
+        elif self.country.lower() == "uk":
+            return "£"
+        else:
+            return "$"  # Default to USD if country not recognized
+
+    def ordinal_suffix(self, n):
+        """Return the ordinal suffix for an integer."""
+        if 10 <= n % 100 <= 20:
+            suffix = 'th'
+        else:
+            suffix = {1: 'st', 2: 'nd', 3: 'rd'}.get(n % 10, 'th')
+        return suffix
+
+    def generate_chart_data(self):
+        categories = [str(i) for i in range(1, 12)]
+        values = [getattr(self, f"decile_{i}") for i in range(1, 12)]
+
+        # Filter out categories and values with zero difference
+        non_zero_data = [(cat, val) for cat, val in zip(categories, values) if val != 0]
+
+        if not non_zero_data:
+            fig = go.Figure()
+            fig.add_annotation(
+                x=0.5,
+                y=0.5,
+                xref="paper",
+                yref="paper",
+                text="No differences to display",
+                showarrow=False,
+                font=dict(size=20)
+            )
+            fig.update_layout(
+                title="Absolute change in household income",
+                xaxis=dict(visible=False),
+                yaxis=dict(visible=False)
+            )
+            return fig
+
+        non_zero_categories, non_zero_values = zip(*non_zero_data)
+
+        # Get currency symbol based on country
+        currency_symbol = self._get_currency_symbol()
+
+        # Generate hover texts with raw impact values and change direction
+        hover_texts = [
+            f"This reform would {self._get_change_direction(val)} the income of households in the {i}{self.ordinal_suffix(int(i))} decile by {currency_symbol}{abs(val):,.1f}"
+            for i, val in zip(non_zero_categories, non_zero_values)
+        ]
+
+        fig = go.Figure()
+
+        values_in_bn = non_zero_values  # No need to convert values
+        colors = [self._get_color(value) for value in non_zero_values]
+
+        # Add bar chart with text formatted with currency symbol
+        fig.add_trace(go.Bar(
+            x=non_zero_categories,
+            y=values_in_bn,
+            marker=dict(color=colors, line=dict(width=1)),
+            width=0.6,
+            text=[f"{currency_symbol}{abs(value):,.1f}" for value in non_zero_values],  # Display values with currency symbol
+            textposition='outside',
+            hovertemplate="<b>Decile %{x}</b><br><br>%{customdata}<extra></extra>",  # Hover shows "Decile {x}"
+            customdata=hover_texts
+        ))
+
+        # Update layout to include currency on y-axis
+        fig.update_layout(
+            yaxis=dict(
+                tickformat=",.0f",  # No decimal places for the y-axis, add thousands separator
+                title=f"Absolute Impact on Income ({currency_symbol})"
+            ),
+            xaxis=dict(
+                title="Income Decile"
+            ),
+            hoverlabel=dict(
+                bgcolor="white",
+                font=dict(color="black", size=16)
+            ),
+            title="Absolute Change in Household Income by Decile"
+        )
+
+        format_fig(fig)  # Keep the formatting logic from policyengine_core
+        return fig

policyengine/charts/distributional_impact/by_income_decile/relative.py

@@ -0,0 +1,104 @@
+import plotly.graph_objects as go
+from policyengine_core.charts.formatting import *
+
+
+class ByIncomeDecileRelativeChart:
+    def __init__(self, country: str, data=None):
+        if data is None:
+            raise ValueError("Data must be provided")
+
+        # Convert the relative values to percentages and store them in attributes for each decile
+        for i in range(1, 12):
+            setattr(self, f"decile_{i}", data['relative'][i] * 100)
+
+        self.country = country
+
+    def _get_color(self, value):
+        if value is None or value == 0 or value < 0:
+            return GRAY
+        return BLUE
+
+    def _get_change_direction(self, value):
+        if value > 0:
+            return "increase"
+        elif value < 0:
+            return "decrease"
+        else:
+            return "no change"
+
+    def ordinal_suffix(self, n):
+        """Return the ordinal suffix for an integer."""
+        if 10 <= n % 100 <= 20:
+            suffix = 'th'
+        else:
+            suffix = {1: 'st', 2: 'nd', 3: 'rd'}.get(n % 10, 'th')
+        return suffix
+
+
+    def generate_chart_data(self):
+        categories = [str(i) for i in range(1, 12)]
+        values = [getattr(self, f"decile_{i}") for i in range(1, 12)]
+
+        # Filter out categories and values with zero difference
+        non_zero_data = [(cat, val) for cat, val in zip(categories, values) if val != 0]
+
+        if not non_zero_data:
+            fig = go.Figure()
+            fig.add_annotation(
+                x=0.5,
+                y=0.5,
+                xref="paper",
+                yref="paper",
+                text="No differences to display",
+                showarrow=False,
+                font=dict(size=20)
+            )
+            fig.update_layout(
+                title="Relative change in household income",
+                xaxis=dict(visible=False),
+                yaxis=dict(visible=False)
+            )
+            return fig
+
+        non_zero_categories, non_zero_values = zip(*non_zero_data)
+
+        # Generate hover texts with formatted impact values and change direction
+        hover_texts = [
+            f"This reform would {self._get_change_direction(val)} the income of households in the {i}{self.ordinal_suffix(int(i))} decile by {abs(val):,.1f}%"
+            for i, val in zip(non_zero_categories, non_zero_values)
+        ]
+
+        fig = go.Figure()
+
+        values_in_bn = non_zero_values  # The values are already in percentages
+        colors = [self._get_color(value) for value in non_zero_values]
+
+        fig.add_trace(go.Bar(
+            x=non_zero_categories,
+            y=values_in_bn,
+            marker=dict(color=colors, line=dict(width=1)),
+            width=0.6,
+            text=[f"{value:.1f}%" for value in non_zero_values],  # Display values with one decimal place and percentage symbol
+            textposition='outside',
+            hovertemplate="<b>Decile %{x}</b><br><br>%{customdata}<extra></extra>",  # Hover shows "Decile {x}"
+            customdata=hover_texts
+        ))
+
+        # Update layout to show percentage on y-axis and format figure
+        fig.update_layout(
+            yaxis=dict(
+                tickformat=".1f%",  # Format for one decimal place with percentage symbol
+                ticksuffix="%",
+                title="Relative Impact on Income (%)"
+            ),
+            xaxis=dict(
+                title="Income Decile"
+            ),
+            hoverlabel=dict(
+                bgcolor="white",
+                font=dict(color="black", size=16)
+            ),
+            title="Relative Change in Household Income by Decile"
+        )
+        format_fig(fig)  # Keep the formatting logic from policyengine_core
+        return fig

policyengine/charts/distributional_impact/by_wealth_decile/average.py

@@ -0,0 +1,114 @@
+import plotly.graph_objects as go
+from policyengine_core.charts.formatting import *
+
+class ByWealthDecileAverageChart:
+    def __init__(self, country: str, data=None):
+        if data is None:
+            raise ValueError("Data must be provided")
+
+        # Store values as they are (no percentage conversion)
+        for i in range(1, 12):
+            setattr(self, f"decile_{i}", data['average'][i])
+
+        self.country = country
+
+    def _get_color(self, value):
+        if value is None or value == 0 or value < 0:
+            return GRAY
+        return BLUE
+
+    def _get_change_direction(self, value):
+        if value > 0:
+            return "increase"
+        elif value < 0:
+            return "decrease"
+        else:
+            return "no change"
+
+    def _get_currency_symbol(self):
+        if self.country.lower() == "us":
+            return "$"
+        elif self.country.lower() == "uk":
+            return "£"
+        else:
+            return "$"  # Default to USD if country not recognized
+
+    def ordinal_suffix(self, n):
+        """Return the ordinal suffix for an integer."""
+        if 10 <= n % 100 <= 20:
+            suffix = 'th'
+        else:
+            suffix = {1: 'st', 2: 'nd', 3: 'rd'}.get(n % 10, 'th')
+        return suffix
+
+    def generate_chart_data(self):
+        categories = [str(i) for i in range(1, 12)]
+        values = [getattr(self, f"decile_{i}") for i in range(1, 12)]
+
+        # Filter out categories and values with zero difference
+        non_zero_data = [(cat, val) for cat, val in zip(categories, values) if val != 0]
+
+        if not non_zero_data:
+            fig = go.Figure()
+            fig.add_annotation(
+                x=0.5,
+                y=0.5,
+                xref="paper",
+                yref="paper",
+                text="No differences to display",
+                showarrow=False,
+                font=dict(size=20)
+            )
+            fig.update_layout(
+                title="Absolute change in household income",
+                xaxis=dict(visible=False),
+                yaxis=dict(visible=False)
+            )
+            return fig
+
+        non_zero_categories, non_zero_values = zip(*non_zero_data)
+
+        # Get currency symbol based on country
+        currency_symbol = self._get_currency_symbol()
+
+        # Generate hover texts with raw impact values and change direction
+        hover_texts = [
+            f"This reform would {self._get_change_direction(val)} the income of households in the {i}{self.ordinal_suffix(int(i))} decile by {currency_symbol}{abs(val):,.1f}"
+            for i, val in zip(non_zero_categories, non_zero_values)
+        ]
+
+        fig = go.Figure()
+
+        values_in_bn = non_zero_values  # No need to convert values
+        colors = [self._get_color(value) for value in non_zero_values]
+
+        # Add bar chart with text formatted with currency symbol
+        fig.add_trace(go.Bar(
+            x=non_zero_categories,
+            y=values_in_bn,
+            marker=dict(color=colors, line=dict(width=1)),
+            width=0.6,
+            text=[f"{currency_symbol}{abs(value):,.1f}" for value in non_zero_values],  # Display values with currency symbol
+            textposition='outside',
+            hovertemplate="<b>Decile %{x}</b><br><br>%{customdata}<extra></extra>",  # Hover shows "Decile {x}"
+            customdata=hover_texts
+        ))
+
+        # Update layout to include currency on y-axis
+        fig.update_layout(
+            yaxis=dict(
+                tickformat=",.0f",  # No decimal places for the y-axis, add thousands separator
+                title=f"Absolute Impact on Wealth ({currency_symbol})"
+            ),
+            xaxis=dict(
+                title="Wealth Decile"
+            ),
+            hoverlabel=dict(
+                bgcolor="white",
+                font=dict(color="black", size=16)
+            ),
+            title="Absolute Change in Household Income by Decile"
+        )
+
+        format_fig(fig)  # Keep the formatting logic from policyengine_core
+        return fig