Portfolio Analytics
Scenario-based portfolio analysis across the Dilikoglu Research coverage universe.
Hypothetical allocation using today’s research inputs; not an investable historical track record. Gross returns exclude trading costs and taxes.
Performance & allocation
Portfolio setup
Attribution & Risk
| Ticker | Weight | Period Return | Linked Contribution | Covariance Risk Share |
|---|
| Metric | Value |
|---|---|
| Tracking Error | — |
| Beta to Benchmark | — |
| Portfolio Sharpe (SOFR) | — |
| Benchmark Sharpe (SOFR) | — |
| Total Rebalance Turnover | — |
| Selected Names | — |
| Rule | — |
Allocation Inputs & Model Provenance
Each run reads the current published probability-weighted intrinsic value. Reference prices are frozen research inputs, not live quotes. Evidence scores reflect Dilikoglu Research analyst judgment. Scores influence the valuation-based rule; other rules use their selected weighting method. Target weights are set at the last rebalance. Ending weights reflect market-price movements since the last rebalance or starting allocation. Equal weight assigns 1/N to every selected holding at each rebalance.
| Company | Model date | Reference price ($) | PW IV ($) | Published score | Applied score | Score basis | Target weight | Ending weight |
|---|
Methodology & sources
Retrospective allocation diagnostic: Current valuations and evidence scores are applied retrospectively. These results are not a historical record of investable recommendations and cannot establish the predictive skill of today’s valuation or evidence scores. A point-in-time research-score backtest is not available. Gross returns exclude trading costs and taxes.
Run the analysis to see the selected method.
Valuation score = clamp(50 + 100 × (1 − reference price / PW IV), 0, 100). Combined score = 70% valuation + 30% evidence. Valuation-based allocation uses score² / √trailing volatility, then normalizes and caps weights at rebalance.
Linked contribution allocates cumulative portfolio return among holdings. Covariance risk share measures each holding’s contribution to estimated portfolio variance. Information ratio compares active return with tracking error. Annualized measures are sensitive to this short sample.
Sources: published company models, adjusted market-price history.
SOFR risk-free reference is shown after analysis.
SOFR source · Federal Reserve Bank of New York
Benchmark context
QQQ is the sole performance benchmark. It tracks the Nasdaq-100 and provides a broad growth comparison as coverage expands across segments; it is not a sector-matched index.
Benchmark methodology · Invesco. Performance uses QQQ market-price history, not a reconstruction of its holdings. The reported period begins when every selected holding and QQQ have usable return history.
Company Overview
Live market data, current valuation anchors and the selected company's published research model.
Live Price vs Frozen Valuation
Price history and the current market price versus the frozen Failure, Bear, Base and Bull underwriting ranges.
Sentiment Analysis
Reddit post sentiment (7-day sample)
We score the available Reddit posts within the seven-day window shown below. Observations update up to every six hours. When newer data are unavailable, an older sample is identified with its observation date. Stocktwits is a separate observation with its own timestamp. The price-chart period does not change sentiment coverage.
- Data currently unavailable
Recent Posts (Reddit)
Data currently unavailable| Date | Channel / author | Post Title / Excerpt | Sentiment | Weight | Engagement | Link |
|---|
Methodology
Important Notes
- The 7D weighted score uses public Reddit posts only.
- Stocktwits is separate; its measurement window is not verified by this dashboard.
- Sentiment is a positioning indicator, not investment advice.
Watchlist
Live pricing, daily moves, market capitalization and 52-week ranges across the monitored universe.
| Ticker | Price | 1D | Market Cap | 52W Low | 52W High |
|---|
Independent research built to be challenged, not marketed.
Dilikoglu Research is an independent research publication focused on transparent assumptions, primary evidence, scenario analysis and capital-aware valuation.
About Tugay Dilikoğlu
Tugay Dilikoğlu is the founder of Dilikoglu Research, an independent equity research platform focused on AI infrastructure and the semiconductor value chain.
His professional background spans treasury product control, valuation, financial forecasting, balance-sheet analysis and financial markets. That experience informs a research approach centered on how operating performance translates into equity value after accounting for capital intensity, financing, dilution and balance-sheet risk.
At Dilikoglu Research, company underwriting is built from primary-source evidence and company-specific operating models. Reported facts, management guidance and independent estimates are treated separately, while valuation is tested through explicit scenarios, sensitivities and thesis falsifiers rather than a single-point forecast.
Tugay holds a BA in Economics from Boğaziçi University and an MSc in Finance from Koç University.
Research published by Dilikoglu Research is produced independently and separately from his professional responsibilities.
Research approach
The objective is not to publish a single-point forecast. Each company is underwritten through explicit Failure, Bear, Base and Bull cases, with assumptions separated from evidence and financing treated as part of the investment case rather than an afterthought.
Published valuations use dated research assumptions and reference prices. Models are reviewed as new filings, contracts, operating data and market evidence change the investment case. Sentiment and market positioning are tracked separately from intrinsic value.
Portfolio Analytics emphasizes discounts to probability-weighted intrinsic value, alongside evidence quality, volatility and position limits. Simulations apply current research scores retrospectively; they are not a record of historical investment decisions.
Connect on LinkedIn
For professional contact, research discussion or feedback on Dilikoglu Research, use LinkedIn.