Spaces:
Sleeping
Sleeping
Download visualization/var.py from MinsaitEU/VaR_model: direct link, hf CLI and curl.
- Browser
- Download file 3.65 kB
-
https://huggingface.co/spaces/MinsaitEU/VaR_model/resolve/main/visualization/var.py
- Command line
-
hf download hf://spaces/MinsaitEU/VaR_model/visualization/var.py
-
curl -L -o var.py https://huggingface.co/spaces/MinsaitEU/VaR_model/resolve/main/visualization/var.py
3.65 kB
| import pandas as pd | |
| import plotly.express as px | |
| import plotly.graph_objects as go | |
| from data.schema import Out | |
| def metrics_dataframe(metrics): | |
| df = pd.DataFrame(metrics) | |
| return df | |
| def plot_loss_distribution( | |
| metrics, | |
| ): | |
| df = pd.DataFrame(metrics) | |
| df[Out.CONFIDENCE] = ( | |
| df[Out.CONFIDENCE] * 100 | |
| ) | |
| fig = px.bar( | |
| df, | |
| x=Out.CONFIDENCE, | |
| y=Out.VAR, | |
| title="VaR by Confidence Level", | |
| text=Out.VAR, | |
| ) | |
| fig.update_layout( | |
| template="plotly_white", | |
| xaxis_title="Confidence Level (%)", | |
| yaxis_title="VaR", | |
| ) | |
| return fig | |
| def plot_var_es( | |
| metrics, | |
| ): | |
| df = pd.DataFrame(metrics) | |
| df[Out.CONFIDENCE] = ( | |
| df[Out.CONFIDENCE] * 100 | |
| ) | |
| fig = go.Figure() | |
| fig.add_trace( | |
| go.Bar( | |
| x=df[Out.CONFIDENCE], | |
| y=df[Out.VAR], | |
| name=Out.VAR, | |
| ) | |
| ) | |
| fig.add_trace( | |
| go.Bar( | |
| x=df[Out.CONFIDENCE], | |
| y=df[Out.ES], | |
| name=Out.ES, | |
| ) | |
| ) | |
| fig.update_layout( | |
| title="VaR vs Expected Shortfall", | |
| xaxis_title="Confidence Level (%)", | |
| yaxis_title="Loss", | |
| barmode="group", | |
| template="plotly_white", | |
| ) | |
| return fig | |
| def plot_component_var(component_metrics, level="Item"): | |
| """ | |
| Bar chart of Component VaR by item (plant, factor, ...). | |
| """ | |
| df = pd.DataFrame(component_metrics) | |
| if df.empty: | |
| raise ValueError( | |
| "No Component VaR results to plot." | |
| ) | |
| confidence = df[Out.CONFIDENCE].iloc[0] * 100 | |
| df = df.sort_values( | |
| Out.COMPONENT_VAR, | |
| ascending=True, | |
| ) | |
| fig = px.bar( | |
| df, | |
| x=Out.COMPONENT_VAR, | |
| y=Out.ITEM, | |
| orientation="h", | |
| text=Out.COMPONENT_VAR, | |
| title=( | |
| f"Component VaR by {level} " | |
| f"({confidence:.1f}% confidence)" | |
| ), | |
| hover_data=[ | |
| Out.STANDALONE_VAR, | |
| Out.CONTRIBUTION, | |
| Out.DIVERSIFICATION, | |
| Out.COMPONENT_ES, | |
| ], | |
| ) | |
| fig.add_vline( | |
| x=0, | |
| line_color="black", | |
| line_width=1, | |
| ) | |
| fig.update_layout( | |
| template="plotly_white", | |
| xaxis_title="Component VaR", | |
| yaxis_title=level, | |
| ) | |
| return fig | |
| def plot_component_var_contribution( | |
| component_metrics, | |
| level="Item", | |
| ): | |
| """ | |
| Bar chart of percentage contribution to portfolio VaR. | |
| Negative contributions can indicate hedging effects. | |
| """ | |
| df = pd.DataFrame(component_metrics) | |
| if df.empty: | |
| raise ValueError( | |
| "No Component VaR results to plot." | |
| ) | |
| confidence = df[Out.CONFIDENCE].iloc[0] * 100 | |
| df = df.sort_values( | |
| Out.CONTRIBUTION, | |
| ascending=True, | |
| ) | |
| fig = px.bar( | |
| df, | |
| x=Out.CONTRIBUTION, | |
| y=Out.ITEM, | |
| orientation="h", | |
| text=Out.CONTRIBUTION, | |
| title=( | |
| f"VaR Contribution by {level} " | |
| f"({confidence:.1f}% confidence)" | |
| ), | |
| hover_data=[ | |
| Out.COMPONENT_VAR, | |
| Out.PORTFOLIO_VAR, | |
| Out.STANDALONE_VAR, | |
| ], | |
| ) | |
| fig.add_vline( | |
| x=0, | |
| line_color="black", | |
| line_width=1, | |
| ) | |
| fig.update_layout( | |
| template="plotly_white", | |
| xaxis_title="Contribution to Portfolio VaR (%)", | |
| yaxis_title=level, | |
| ) | |
| return fig |