SNSaranne Ndamba Download model

Investment research project · Prices as of 17 September 2026

Growth, without
unpriced conviction.

A €100,000 global portfolio built to capture structural earnings growth while controlling concentration, valuation and drawdown risk.

Designed and modelled by Saranne Ndamba · MSc Finance & Investment Management
Three-year outcome
Base case value · Year 3
€139,423
+39.4% cumulative return on €100,000
€100k invested
TodayYear 1Year 2Year 3
Capital€100,000
Expected Y3 value€132,126
Base CAGR11.7%
Est. volatility17.2%
Weighted beta0.89

01 / Portfolio construction

One portfolio. Three possible futures.

Switch the scenario to see the three-year outcome and how each position contributes to it.

Target allocation · 16 positions
Equity · 10 companies 73%Fixed income 12%Gold & infrastructure 7%Put-spread hedge 3%Cash 5%
Base case · 3-year outcome
€139,423
Annualised return+11.7%
Total gain / loss+€39,423
Scenario probability55%
Probability-weighted value€132,126 +32.1%
Earnings compound as expected while valuation multiples remain broadly stable.
Construction rules
  1. 01Growth must convert into free cash flow.
  2. 02The screen score sets each position's maximum weight: 12%, 10% or 8%.
  3. 03Every holding has a measurable exit trigger.
  4. 04Diversifiers must have an explicit portfolio role.

02 / Equity research

Conviction is earned company by company.

Filter the holdings, compare scenario returns, and open any position to see its valuation screen, conviction tier, case and risks.

SecurityThemeWeightBase CAGRYear 3 valueDetails

The headline outcome assumes positions are rebalanced back to target weights each year, so winners are trimmed. Left to drift, the base case would end €982 higher, with more concentration risk. That gap is the cost of the rebalancing discipline.

03 / Risk management

Risk is designed, not discovered later.

Every risk is tied to an early-warning trigger and a set response.

17.2%

Estimated volatility

Holding-level volatilities with a simplified 0.30 common correlation keep the portfolio inside its 15–18% design range.

Factor control

Weighted beta is 0.89 against a 1.0 ceiling. High-beta positions are reduced first when risk rises.

Rebalancing

Reviewed quarterly, or sooner when a position drifts 25% from its target weight.

Tail protection

A 3% annual put-spread budget caps premium drag while adding convex protection in a severe sell-off.

RiskEarly warningPortfolio response
Position drift±25% from target weightTrade back toward target
Valuation compressionBase-case return falls below 7%Trim or replace
Thesis impairmentTwo consecutive operating missesRe-underwrite; exit if structural
Ethical breachMaterial governance, labour or human-rights evidenceEscalate, engage or exclude
Hedge expiryPut spread expires or equity weight changesRoll and resize within the 3% budget

04 / Analytical framework

From CFA concepts to portfolio decisions.

Nine disciplines applied directly, each with a decision rule you can check in the model.

05 / Risk lab

Measured three ways, then checked against reality.

Python (numpy, pandas) on 10,000 simulated paths and five years of real daily prices, with the same VaR rebuilt in Excel.

Realised volatility19.6%Model assumption 17.2%
1-day VaR, 99%€3,328Historical, on €100,000
Expected Shortfall, 99%€3,911Average loss beyond the VaR
Maximum drawdown−25.4%Trough on 28 Dec 2022
Sharpe ratio0.56Expected; 0.91 realised
Value at Risk and Expected Shortfall · €100,000
MethodVaR 95%ES 95%VaR 99%ES 99%

Distribution of outcomes
OutcomesTail beyond VaR 95%VaR 95% and 99%
Monte Carlo · 10,000 paths of €100,000 over one year
5th to 95th percentile25th to 75th percentileMedian pathSample paths
Backtest · today's weights on real prices, Sep 2021 to Sep 2026

A backtest of the current portfolio benefits from hindsight: these companies were chosen knowing how the last five years went. Read the return as an illustration, and the volatility, drawdown and tail losses as the useful evidence.

Model check · assumed against realised volatility
Assumed in the modelRealised, 5 years
What reality said about my model
    Where the risk comes from · share of portfolio volatility

    Value at Risk valeur en risque (VaR)

    The loss that should not be exceeded at a given confidence over a given horizon. A 1-day 95% VaR of €2,025 means that on 19 days out of 20, the portfolio loses less than €2,025.

    Expected Shortfall perte moyenne au-delà de la VaR (CVaR)

    The average loss on the days that are worse than the VaR. Basel's market risk rules (FRTB) replaced VaR with it because VaR says nothing about how deep the tail goes.

    Three methods paramétrique, historique, Monte-Carlo

    Parametric assumes normal returns and scales volatility (1.645 at 95%, 2.326 at 99%). Historical replays real days on today's weights. Monte Carlo draws 10,000 correlated scenarios from the model's volatilities and correlations.

    Fat tails and backtesting queues épaisses, backtesting

    Real markets have more extreme days than a normal curve predicts. Counting how often losses beat the VaR (a Kupiec-style test) shows whether the model is too optimistic.

    Sharpe ratio ratio de Sharpe

    Return above the risk-free rate, divided by volatility: the reward for each unit of risk. Here the risk-free rate is set at 2.0% (euro short-term rate).

    Maximum drawdown perte maximale depuis un sommet

    The worst fall from a peak to a later trough. It is the number clients feel, and the one that decides whether they stay invested.

    Method: 16 holdings and weights from the Excel model. Daily closing prices from Google Finance, 17 Sep 2021 to 17 Sep 2026, price returns without dividends. US-listed holdings are translated into euros at the daily EUR/USD rate. Bond, gold and infrastructure sleeves use liquid proxies (AGG, LQD, GLD, IGF); the put-spread budget is represented by −1.5× the S&P 500 daily move less its premium. Monte Carlo uses the model's own assumptions (base-case returns, holding volatilities, 0.30 common correlation), seed 2026. Risk-free rate 2.0%.

    06 / Regulation check

    Could it be sold as a UCITS fund? Not yet. Here is the fix.

    The portfolio tested against the UCITS diversification rules, its PRIIPs risk indicator computed as in a real KID, and what would change under AIFMD.

    UCITS diversification · Directive 2009/65/EC, articles 52 and 55
      The 5/10/40 rule · issuers above 5%

      A UCITS may put up to 10% in one issuer, but the issuers above 5% must add up to 40% or less. ETFs that are themselves UCITS follow a separate 20% limit per fund.

      Holding by holding
      CompliantCounts toward 40% or checkBreach
      HoldingWeightUCITS treatment
      The fix · same 73% in equities, now compliant
      PositionNowCompliantChange

      Trim the five largest names so that no issuer exceeds 10% and the group above 5% totals exactly 40%. The 9% released goes into a global equity UCITS ETF, which falls under the separate 20% per-fund limit. The equity allocation stays at 73%, conviction stays in the largest names, and model volatility falls.

      The ETF line (IWDA, a physically replicated global equity UCITS ETF) is an illustrative choice; any broad equity UCITS within the 20% limit works.

      Key Information DocumentPRIIPs · illustrative

      What are the risks and what could I get in return?

      Risk indicator
      ← Lower riskHigher risk →

      Be aware of currency risk. Most holdings are priced in US dollars, so the final return depends on the exchange rate between the dollar and the euro. This risk is not considered in the indicator shown above.

      This product does not include any protection from future market performance, so you could lose some or all of your investment.

      How the indicator is calculated · PRIIPs Regulation, Annex II

      The same portfolio as an AIF · what changes under AIFMD
      Luxembourg view
      As a UCITSAs an AIFFor this portfolio
      InvestorsRetail, with a European marketing passportMainly professional investors; retail only where national rules allowUCITS fits a fund sold to individuals
      Diversification5/10/40 rule, 20% per fund, 20% per bankSet by the fund's own documents (a SIF or RAIF must still spread risk, typically no more than 30% in one issuer)NVIDIA at 12% and the 59% group would be allowed
      LeverageGlobal exposure capped at 100% of NAV (commitment), or absolute VaR at 20% of NAV (99%, 20 days)Measured by the gross and commitment methods and disclosed; above 3 times NAV is "substantial" and reported to the regulator, which can impose limits
      Eligible assetsListed securities, money market instruments, UCITS funds, deposits, financial derivativesAlmost anything the strategy allows: physical gold, private equity, real estate, loansGold could be held directly instead of through an ETC
      LiquidityRedemptions at least twice a month, usually dailyOpen or closed-ended; notice periods, gates and lock-ups possibleBoth regimes now require liquidity management tools (AIFMD II and the UCITS amendments, from April 2026)
      DepositaryMandatory: safekeeping, cash monitoring, oversightAlso mandatory under AIFMD, with the same three duties and strict liability for lost assetsNo difference in principle
      ReportingKID (PRIIPs), prospectus, annual and semi-annual reports, risk reporting to the CSSFAnnex IV report to the regulator (quarterly to annual, by size), investor disclosures, KID only if sold to retailAn AIF adds Annex IV: leverage, liquidity, exposures and stress tests
      Luxembourg vehiclePart I UCITS under the Law of 17 December 2010 (SICAV or FCP)Part II UCI, SIF (2007 Law) or RAIF (2016 Law), managed by an authorised AIFMA RAIF launches fastest: no CSSF product approval, oversight sits with the AIFM

      UCITS OPCVM

      The EU's harmonised retail fund. Strict rules on diversification, eligible assets, leverage and liquidity buy a passport to sell it across Europe. Luxembourg and Ireland domicile most of them.

      The 5/10/40 rule la règle des 5/10/40

      At most 10% in one issuer, and the issuers above 5% together at most 40%. It forces at least 16 meaningful positions and stops a fund becoming a bet on a few names.

      PRIIPs KID and SRI DIC PRIIPs, indicateur synthétique de risque (ISR)

      Since 2023 every UCITS sold to retail has a KID. Its 1 to 7 risk score comes from the VaR-equivalent volatility over the holding period, calculated with skew and fat tails (Cornish-Fisher).

      AIF and AIFMD FIA, directive AIFM

      Every EU fund that is not a UCITS is an AIF. The directive regulates the manager (the AIFM) rather than the product: authorisation, leverage, liquidity, depositary and reporting.

      Global exposure risque global, méthode de l'engagement

      How much market exposure derivatives add. Under the commitment approach a UCITS can add at most 100% of NAV; the VaR approach caps absolute VaR at 20%.

      Depositary dépositaire

      An independent bank that holds the assets, monitors the cash and checks the manager follows the rules. If it loses a security, it must replace it.

      Methodology

      A transparent, auditable process.

      Each holding has explicit bear, base and bull annual-return assumptions, applied to point-in-time prices from 17 September 2026. Three-year values use compound growth with annual rebalancing to target weights. Scenario probabilities (25% / 55% / 20%) and volatility inputs are judgemental and disclosed as such, not presented as facts.

      US-listed prices are translated into euros at 1.18 USD per EUR, and currency is not separately forecast. Taxes, spreads, custody costs and option execution are excluded. This is an investment-research demonstration, not personalised investment advice.

      Download the Excel model (7 sheets)
      Model checks6 / 6
      • ✓Weights reconcile to 100%
      • ✓Capital allocation ties to €100,000
      • ✓Bear, base and bull cases are explicit
      • ✓Every weight sits within its conviction-tier cap
      • ✓Financial and ethical screens included
      • ✓Risk limits, sources and limitations disclosed