report_builders/html_risk.py

import numpy as np

def compute_var(returns, alpha=0.95):
    return float(-np.percentile(returns, (1.0 - alpha) * 100.0))

def compute_cvar(returns, alpha=0.95):
    cutoff = np.percentile(returns, (1.0 - alpha) * 100.0)
    tail_losses = -returns[returns < cutoff]
    if len(tail_losses) == 0:
        return compute_var(returns, alpha)
    return float(np.mean(tail_losses))

def build_cvar_garch_html(returns_df, w_risky, cfg, model_info):
    cvar_garch_html = ""
    _rm = returns_df.values if returns_df is not None else None
    _wa = w_risky.reindex(returns_df.columns).fillna(0.0).values if returns_df is not None else None
    
    if _rm is not None and (cfg.get('cvar_enabled') or cfg.get('garch_enabled')):
        cvar_rows = ""; garch_rows = ""
        
        if cfg.get('cvar_enabled'):
            _a = cfg.get('cvar_alpha', 0.95)
            _port_rets = _rm @ _wa
            cvar_rows = (
                f"<tr><td>1-Day VaR ({_a:.0%})</td>"
                f"<td style='color:#e3b341'>{compute_var(_port_rets, _a):.2%}</td></tr>"
                f"<tr><td>1-Day CVaR ({_a:.0%})</td>"
                f"<td style='color:#f85149'>{compute_cvar(_port_rets, _a):.2%}</td></tr>"
            )
            
        gi = model_info.get('garch_info', {})
        if cfg.get('garch_enabled') and gi:
            for t, gd in gi.items():
                if t == 'CASH': continue
                ratio = gd['garch_ann_vol'] / max(gd['hist_ann_vol'], 0.001)
                rc = "#f85149" if ratio > 1.15 else ("#3fb950" if ratio < 0.90 else "#e3b341")
                badge = "▲ Elevated" if ratio > 1.15 else ("▼ Compressed" if ratio < 0.90 else "≈ Normal")
                garch_rows += (
                    f"<tr><td><strong>{t}</strong></td>"
                    f"<td>{gd['hist_ann_vol']:.1%}</td>"
                    f"<td style='color:{rc}'>{gd['garch_ann_vol']:.1%}</td>"
                    f"<td>{gd['persistence']:.3f}</td>"
                    f"<td style='color:{rc}'>{badge}</td></tr>\n"
                )
                
        if cvar_rows or garch_rows:
            cvar_garch_html = '<p class="st">CVaR &amp; GARCH Volatility Regime</p><div class="two">'
            if cvar_rows:
                cvar_garch_html += (
                    '<div class="cc"><p style="font-size:.8rem;color:#8b949e;margin-bottom:8px">'
                    '<strong>Tail Risk</strong></p>'
                    f'<table><tbody>{cvar_rows}</tbody></table></div>'
                )
            if garch_rows:
                cvar_garch_html += (
                    '<div class="cc" style="overflow-x:auto">'
                    '<p style="font-size:.8rem;color:#8b949e;margin-bottom:8px">'
                    '<strong>GARCH(1,1) Forecasts</strong></p>'
                    '<table><thead><tr><th>Ticker</th><th>Hist</th><th>GARCH</th>'
                    f'<th>Persist.</th><th>Regime</th></tr></thead>'
                    f'<tbody>{garch_rows}</tbody></table></div>'
                )
            cvar_garch_html += '</div>'
            
    return cvar_garch_html

def build_risk_attr_html(mvar_series, cvar_components, factor_ret_attr):
    risk_attr_html = ""
    _has_risk_attr = any(x is not None for x in [mvar_series, cvar_components, factor_ret_attr])
    if _has_risk_attr:
        ra_parts = []
        if mvar_series is not None and not mvar_series.empty:
            mvar_rows = ""
            total_mvar = float(mvar_series.abs().sum()) or 1.0
            for ticker, mv in mvar_series.sort_values(ascending=False).items():
                mv_f    = float(mv)
                mv_col  = "#f85149" if mv_f > 0 else "#3fb950"
                pct_bar = abs(mv_f) / total_mvar
                bar_w   = round(pct_bar * 80, 1)   
                mvar_rows += (
                    f"<tr><td><strong>{ticker}</strong></td>"
                    f"<td style='color:{mv_col};font-weight:700'>{mv_f*100:+.3f}%</td>"
                    f"<td><div style='background:{mv_col};width:{bar_w}px;height:8px;border-radius:2px;opacity:.75'></div></td></tr>\n"
                )
            ra_parts.append(
                '<div class="cc" style="overflow-x:auto"><p style="font-size:.8rem;color:#8b949e;margin-bottom:8px">'
                '<strong>Marginal VaR (95%)</strong> &nbsp;<span style="font-weight:400">— how much each asset adds to portfolio VaR for a 1% increase in its weight</span></p>'
                f'<table><thead><tr><th>Ticker</th><th>MVaR</th><th>Relative Size</th></tr></thead><tbody>{mvar_rows}</tbody></table></div>'
            )
        
        if cvar_components is not None:
            comp_series, total_cvar = cvar_components
            if hasattr(comp_series, 'items') and total_cvar > 0:
                cvar_rows = ""
                for ticker, cv in comp_series.sort_values(ascending=False).items():
                    cv_f    = float(cv)
                    share   = cv_f / total_cvar if total_cvar else 0.0
                    cv_col  = "#f85149" if cv_f > 0 else "#3fb950"
                    bar_w   = round(abs(share) * 80, 1)
                    cvar_rows += (
                        f"<tr><td><strong>{ticker}</strong></td>"
                        f"<td style='color:{cv_col};font-weight:700'>{cv_f*100:+.3f}%</td>"
                        f"<td style='color:#8b949e'>{share:.1%}</td>"
                        f"<td><div style='background:{cv_col};width:{bar_w}px;height:8px;border-radius:2px;opacity:.75'></div></td></tr>\n"
                    )
                ra_parts.append(
                    '<div class="cc" style="overflow-x:auto"><p style="font-size:.8rem;color:#8b949e;margin-bottom:8px">'
                    f'<strong>CVaR Component Attribution</strong> &nbsp;<span style="font-weight:400">— each asset\'s contribution to the portfolio\'s expected loss on its worst days (total CVaR: <span style="color:#f85149">{total_cvar*100:.2f}%</span>)</span></p>'
                    f'<table><thead><tr><th>Ticker</th><th>Component CVaR</th><th>Share of Total</th><th>Relative Size</th></tr></thead><tbody>{cvar_rows}</tbody></table></div>'
                )
                
        if factor_ret_attr:
            fat_rows = ""
            tot_ret = factor_ret_attr["total_return"]
            
            fat_rows += f"<tr><td><strong>Total Return</strong></td><td colspan='2' style='font-weight:700'>{tot_ret*100:+.2f}%</td></tr>"
            
            for f_name, f_val in factor_ret_attr["factor_contributions"].items():
                f_col = "#3fb950" if f_val > 0 else "#f85149"
                fat_rows += f"<tr><td>{f_name} Contribution</td><td style='color:{f_col}'>{f_val*100:+.2f}%</td><td></td></tr>"
            
            alpha_val = factor_ret_attr["alpha"]
            a_col = "#3fb950" if alpha_val > 0 else "#f85149"
            fat_rows += f"<tr><td><strong>Alpha (Idiosyncratic)</strong></td><td style='color:{a_col};font-weight:700'>{alpha_val*100:+.2f}%</td><td></td></tr>"
            
            ra_parts.append(
                '<div class="cc" style="overflow-x:auto"><p style="font-size:.8rem;color:#8b949e;margin-bottom:8px">'
                f'<strong>Factor-based Return Attribution (BHB Proxy)</strong></p>'
                f'<table><thead><tr><th>Factor</th><th>Contribution to Return</th><th></th></tr></thead><tbody>{fat_rows}</tbody></table></div>'
            )
                
        if ra_parts:
            risk_attr_html = '<p class="st">Risk Attribution</p>' + "\n".join(ra_parts)
            
    return risk_attr_html

def build_rc_html(model_info, weights, chart_data):
    rc_html = ""
    rc_series = model_info.get("risk_contributions")
    if rc_series is not None and not rc_series.empty:
        is_ml = ("XGBoost" in model_info.get('name', ''))
        rc_note = (" — <em>Machine Learning Asset Sizing</em>" if is_ml else "")
        rc_rows = ""
        for t in rc_series.index:
            rc_pct = float(rc_series[t])*100
            w_pct = float(weights.get(t,0))*100
            diff = rc_pct - w_pct
            d_col = "#f85149" if diff > 5 else ("#e3b341" if diff > 1 else "#3fb950")
            rc_rows += (
                f"<tr><td><strong>{t}</strong></td>"
                f"<td>{w_pct:.1f}%</td>"
                f"<td style='color:{d_col}'>{rc_pct:.1f}%</td>"
                f"<td style='color:{d_col}'>{diff:+.1f}%</td></tr>\n"
            )
        rc_html = (
            f'<p class="st">Risk Contribution{rc_note}</p>'
            '<div class="two"><div class="cc" style="overflow-x:auto">'
            '<table><thead><tr><th>Ticker</th><th>Weight</th>'
            '<th>Risk Contrib.</th><th>Δ</th></tr></thead>'
            f'<tbody>{rc_rows}</tbody></table></div>'
            '<div class="cc"><canvas id="cj_rc"></canvas></div></div>'
        )
        
        # Inject RC Chart Data into Chart.js
        _r = lambda x: round(x, 4)
        rc_labels = list(rc_series.index)
        rc_colors = ["#f85149" if (float(rc_series[t])*100 - float(weights.get(t,0))*100) > 5 else ("#e3b341" if (float(rc_series[t])*100 - float(weights.get(t,0))*100) > 1 else "#3fb950") for t in rc_labels]
        chart_data["rc_labels"] = rc_labels
        chart_data["rc_ds"] = [
            {"label":"Risk Contrib %","data":[_r(float(rc_series[t])*100) for t in rc_labels],"backgroundColor":rc_colors,"borderRadius":4},
            {"label":"Weight %", "data":[_r(float(weights.get(t,0))*100) for t in rc_labels],"backgroundColor":"rgba(88,166,255,0.4)","borderRadius":4}
        ]
        
    return rc_html