Example Active ETFs
ETF Concepts
Four example active ETF designs
The following examples show how the EPT-HiFuPi computational allocation method could support documented active ETF design concepts using defined asset universes, holding constraints, turnover controls, daily fee assumptions, transaction-cost assumptions, and immediate dividend reinvestment. They are backtested results, not live funds or recommendations.
These four example portfolios contain only stocks and no bonds. The purpose is to examine active portfolio construction from the equity constituents that make up a market portfolio, such as the S&P 500. These active stock portfolios can then be combined with bonds or other assets in a later multi-step portfolio-combination process.
The four portfolios are intentionally free from exogenous variables so the allocation method can be observed, inspected, and verified directly. Exogenous variables may later be added to any single-step or multi-step portfolio process to create a more tailored portfolio design for a specific firm, mandate, or asset-allocation objective.
Asset universe and market portfolio construction
The initial investable universe contains 155 assets selected from the highest-capitalized stocks dating back to January 2007. Each stock must pass a data-integrity review using three independent data brokers before it is included in the universe. The purpose of this review is to reduce dependence on any single data source and to improve confidence in historical price, return, and corporate-action treatment.
Survivorship bias is addressed by creating a comparison market portfolio named MARK155a. MARK155a is then compared against SPY, the S&P 500 ETF, as a practical survivorship-bias check for the high-capitalization stock universe. The current research results indicate that the impact of survivorship bias appears to be very minimal, if present, for this high-capitalization stock setting. This is consistent with the idea that large-cap companies are less likely to fail abruptly and more likely to be absorbed into, merged with, or replaced by other large-cap companies.
Using the same research approach, a broader market portfolio named MARK255b will also be created. MARK255b extends the concept from 155 high-capitalization stocks to 255 high-capitalization stocks while preserving the same emphasis on data integrity, market comparison, and disciplined portfolio construction.
All market portfolios and all EPT example portfolios are constructed so that dividends are reinvested immediately back into the portfolio. This keeps the comparisons aligned around total-return behavior rather than price-only behavior, and it allows market portfolios and EPT portfolios to be evaluated using a consistent reinvestment assumption.
155-stock universe
Selected from the highest-capitalized stocks dating back to January 2007 and screened through a multi-source data-integrity process.
Three-broker data test
Each included asset must satisfy a review using three independent data brokers to reduce reliance on a single historical data source.
MARK155a comparison
A market portfolio named MARK155a is used as the primary reference portfolio and is compared against SPY to evaluate survivorship-bias impact.
All market portfolios have a 0.5% annual fee imposed on a daily basis, equal to 0.0019841% per day. All portfolios below are constructed using the EPT-HiFuPi computational allocation method, have a 1% annual fee applied on a daily basis, include 10 basis points per transaction applied during reallocations, and assume dividends are reinvested immediately back into the portfolio.
Example 01
Mu8T5a
The universe of assets is the same 155 stocks used in the MARK155a portfolio.
Design elements
- Reallocated monthly
- Holding constraint of 8% at the beginning of each holding period
- Turnover limited to 5% of the entire portfolio during any one reallocation
- Uses the MARK155a asset universe
- Constructed using the EPT-HiFuPi computational allocation method
- Assumes dividends are reinvested immediately back into the portfolio
- Includes a 1% annual fee applied daily
- Includes 10 bps per transaction cost during reallocations
Implementation note
Because the holding constraint is imposed only during reallocations, the portfolio may drift above the 8% holding limit between reallocations.
Example 02
Du8Tp75a
The universe of assets is the same 155 stocks used in the MARK155a portfolio.
Design elements
- Reallocated daily
- Holding constraint of 8% at the beginning of each holding period
- Turnover limited to 0.75% of the entire portfolio during any one reallocation
- Uses the MARK155a asset universe
- Constructed using the EPT-HiFuPi computational allocation method
- Assumes dividends are reinvested immediately back into the portfolio
- Includes a 1% annual fee applied daily
- Includes 10 bps per transaction cost during reallocations
Implementation note
Unlike the monthly version, the portfolio’s holdings are not allowed to drift over several days because the portfolio is reallocated daily.
Example 03
Mu8T5b
The universe of assets is the same 255 stocks used in the MARK255b portfolio.
Design elements
- Reallocated monthly
- Holding constraint of 8% at the beginning of each holding period
- Turnover limited to 5% of the entire portfolio during any one reallocation
- Uses the MARK255b asset universe
- Constructed using the EPT-HiFuPi computational allocation method
- Assumes dividends are reinvested immediately back into the portfolio
- Includes a 1% annual fee applied daily
- Includes 10 bps per transaction cost during reallocations
Implementation note
Because the holding constraint is imposed only during reallocations, the portfolio may drift above the 8% holding limit between reallocations.
Example 04
Du8Tp75b
The universe of assets is the same 255 stocks used in the MARK255b portfolio.
Design elements
- Reallocated daily
- Holding constraint of 8% at the beginning of each holding period
- Turnover limited to 0.75% of the entire portfolio during any one reallocation
- Uses the MARK255b asset universe
- Constructed using the EPT-HiFuPi computational allocation method
- Assumes dividends are reinvested immediately back into the portfolio
- Includes a 1% annual fee applied daily
- Includes 10 bps per transaction cost during reallocations
Implementation note
Unlike the monthly version, the portfolio’s holdings are not allowed to drift over several days because the portfolio is reallocated daily.
Common architecture across the examples
Each example can be described using the same research pipeline: define the universe, verify the data, apply immediate dividend reinvestment, impose holding and turnover constraints, solve the portfolio problem using the EPT-HiFuPi computational allocation method, apply the fee and transaction-cost assumptions, backtest the result, and review portfolio behavior over time.
Universe
MARK155a or MARK255b defines the eligible assets.
Data integrity
Assets are reviewed using multiple independent data brokers.
Constraints
Holding limits and turnover limits define implementation discipline.
EPT-HiFuPi allocation
Each example is constructed with the EPT-HiFuPi computational allocation method and evaluated against market portfolio references and historical results.
Required 30-day verification period
Caution: the application results are designed to be verified
We understand the nature of what firms may see in the server applications. For this reason, 5thN Financial offers a 30-day verification period so that the firm can independently examine and test the computations before entering into an agreement. In fact, 5thN Financial will not enter into an agreement without this 30-day verification period.
During the verification period, 5thN Financial will make available a verification server that can be used to verify the portfolio computations. Each reallocation period will generate an Excel spreadsheet that may be inspected by the firm. That spreadsheet may then be uploaded to the verification server, where an independent computation will be performed and the resulting portfolio weights will be returned.
The verification server is made available without charge during this 30-day verification period. Due to time limitations, 5thN Financial will need to begin charging for this verification phase beginning in 2027.
The returned weights should be compared against the overall weights in the final spreadsheet. It is strongly suggested that the firm also perform hand calculations using the reported weights and prices to verify the overall portfolio value.
The verification server can also accept price or return data directly from an Excel spreadsheet. This allows ticker symbols and dates to be obfuscated while still permitting the computational process, resulting weights, and portfolio-value calculations to be examined.
External app suite
This advisor-restricted site remains static while each analytics module launches on a separate restricted server. These pages are ready as polished placeholders for the apps as they are built out.
Wealth
Portfolio path and wealth-outcome exploration.
Performance
Return summaries, benchmark context, and design diagnostics.
Compare
Side-by-side review of ETF design results, constraints, and assumptions.
Underwater
Drawdown depth, recovery time, and stress-period review.
Withdraw
Withdrawal paths, sequence risk, and sustainability views.
The names and descriptions above are examples only. They are not investment recommendations, securities offerings, fund advertisements, or guarantees of future results.