Model Optimization

← US equity strategies

skfolio-powered allocators across US equities and global index universes. Click any model to open its dedicated page.

Portfolio optimization comparing Max Sharpe, Min Variance, Equal Weight, Inverse Volatility, Max Diversification, and Risk Parity — with equity curves, drawdowns, rolling Sharpe, return distributions, and weights.

Generated Tuesday, Jul 7, 2026

Methodology

Built with the skfolio library. US portfolio uses a subset of S&P 500 stocks; global portfolio uses major equity indices. Models are fitted on historical daily returns — transaction costs not modeled.

US Portfolio (US equity)

80 assets · Select a model to view metrics and plots

Risk–return (all models)

What this shows: Model-level risk/return scatter: volatility on x-axis, total return on y-axis, bubble size by Sharpe.

How to read it: Higher and leftward points with larger bubbles are generally better risk-adjusted candidates.

All models: cumulative returns (wealth index)

What this shows: Overlayed wealth-index curves for all optimization models.

How to read it: Compare long-run slope and drawdown/recovery behavior to spot robust models rather than single-period winners.

All models: drawdown % comparison
Sharpe Ratio
0.97
Volatility %
15.41%
Max Drawdown %
18.34%
Total Return %
29.70%
Sortino Ratio
1.37
Calmar Ratio
0.00
win_rate_pct
53.57%
var_95_pct
1.41%
cvar_95_pct
2.16%
average_drawdown_pct
2.14%
annualized_return_pct
15.00%
Cumulative returns (wealth index)

What this shows: Selected model wealth-index path over time (Equal Weight).

How to read it: Steadier ascent with smaller interruptions is typically preferable to highly jagged paths with similar endpoint return.

Drawdown %
Portfolio weights (top 20)
Rolling Sharpe ratio (21-day, annualized)
Rolling volatility (21-day, annualized %)
Daily returns distribution (%)

Global Portfolio (global indices)

20 assets · Select a model to view metrics and plots

Risk–return (all models)

What this shows: Model-level risk/return scatter: volatility on x-axis, total return on y-axis, bubble size by Sharpe.

How to read it: Higher and leftward points with larger bubbles are generally better risk-adjusted candidates.

All models: cumulative returns (wealth index)

What this shows: Overlayed wealth-index curves for all optimization models.

How to read it: Compare long-run slope and drawdown/recovery behavior to spot robust models rather than single-period winners.

All models: drawdown % comparison
Sharpe Ratio
1.87
Volatility %
9.74%
Max Drawdown %
10.61%
Total Return %
42.42%
Sortino Ratio
2.46
Calmar Ratio
0.01
win_rate_pct
61.11%
var_95_pct
0.84%
cvar_95_pct
1.46%
average_drawdown_pct
0.91%
annualized_return_pct
18.21%
Cumulative returns (wealth index)

What this shows: Selected model wealth-index path over time (Equal Weight).

How to read it: Steadier ascent with smaller interruptions is typically preferable to highly jagged paths with similar endpoint return.

Drawdown %
Portfolio weights (top 20)
Rolling Sharpe ratio (21-day, annualized)
Rolling volatility (21-day, annualized %)
Daily returns distribution (%)
QuantifiedTrader logoQuantifiedTrader

Independent quantitative research on trading methods, backtesting, and market analytics.

Research disclaimer

QuantifiedTrader is operated by an independent quantitative research group. We study, document, and compare different methods of trading, portfolio construction, risk management, and investment analysis. Our work is exploratory and academic in nature—we build tools, run backtests, and publish findings to advance understanding, not to promote any particular strategy or product.

Not investment advice. Nothing on this website constitutes investment, trading, financial, tax, legal, or other professional advice. We do not recommend, endorse, or solicit the purchase or sale of any security, derivative, or financial instrument, nor do we suggest that any strategy, model, or result presented here is suitable for any individual or institution. Any examples, simulations, or performance figures are illustrative research outputs only.

No client or advisory relationship. We do not provide investment advisory, brokerage, portfolio-management, custody, or asset-management services to any person or entity. Browsing this site, using our tools, or contacting us does not create a client, fiduciary, or advisory relationship. We do not manage money on behalf of third parties and do not act as agents for any financial institution.

Research & education only. Content, datasets, backtests, charts, code, and software made available here are for informational and educational research. Materials may be incomplete, simulated, hypothetical, or derived from third-party sources that we do not control. Past performance, backtested results, and historical analyses are not indicative of future results. Market conditions change; models may fail; assumptions may be wrong. You are solely responsible for evaluating any information and for all decisions you make.

No responsibility or liability. To the fullest extent permitted by applicable law, QuantifiedTrader and its contributors disclaim all responsibility and liability for any loss, damage, cost, or expense—direct or indirect—arising from access to, use of, or reliance on this website, its content, or its tools. All materials are provided “as is” and “as available,” without warranties of any kind, whether express or implied, including but not limited to accuracy, completeness, fitness for a particular purpose, or non-infringement.

Non-commercial research sharing. This site does not aim to profit from the knowledge, tools, or datasets published here. Materials are shared for non-commercial research and learning, subject to applicable open-source or site terms where noted. We are a research collective, not a commercial product or service provider.

Contact. For questions about this notice, the site, or published research materials, contact support@quantedx.com. Correspondence is for administrative and research purposes only and does not constitute advice or create any professional obligation on our part.

© 2026 QuantifiedTrader. All rights reserved.