Simple mode: type one ticker and press Enter or click Analyze everything. Noweir runs market data, technicals, confirmed chart patterns, Market Memory, Evidence Graph, fundamentals, company news, world-news mapping, SEC filings, events, portfolio concentration, ETF overlap and stress testing. Then it automatically generates the final decision.
Type a ticker or company name. Use ↑ ↓ and Enter to choose.
1Market data
Price + freshness
2Technical evidence
Patterns + memory + conflicts
3Fundamentals
Valuation + growth + quality
4Catalysts
News + filings + events
5Portfolio fit
Concentration + overlap + stress
6Final decision
Synthesis + blockers
Ready. Enter a stock to begin.
Review detailed gates or complete missing personal inputs
Noweir Decision Engine ?
All tools → one decision packet
Automatic synthesis: every one-click analysis ends here. The engine combines technicals, confirmed patterns, Market Memory, Evidence Graph, fundamentals, catalysts, filings, world-news mapping, dividend/ex-dividend effects, portfolio risk and psychology into one answer. BUY NOW appears only when the configured evidence rules and required personal inputs are satisfied.
Model Entry Intelligence
Preferred entry zone from support, moving averages, ATR, RSI, confirmed patterns, Market Memory, evidence conflicts and news/event risk. Analyst targets stay secondary context.
Lower entry
—
Model midpoint
—
Upper entry
—
Waiting for analysis. Simple mode uses the Balanced profile.
20D trend
—
20D / 63D vol
—
MA alignment
—
63D range position
—
Shape hypotheses
—
Quantitative chart structure will appear after analysis.
Why this zone
Risk adjustments
Historical Calibration
Tests a point-in-time technical proxy across historical daily data, then compares its forward outcomes with the stock's unconditional baseline.
Raw / independent 1M signals
—
Median next 1M
—
Positive next 1M
—
Vs baseline
—
Current setup
—
Waiting for analysis. Historical calibration is descriptive; it does not turn past returns into a future probability.
Independent win rate
—
Profit factor
—
Payoff ratio
—
Median MFE / MAE
—
Sample quality
—
Transaction-cost stress will appear after analysis.
Recent historical signals
Date
Entry
Next 5D
Next 1M
Next 3M
21D best/worst excursion
Dividend & Ex-Dividend Impact
Separates the ex-dividend date from the payment date and checks how this stock actually behaved around recent ex-dividend sessions.
Reported ex-dividend
—
Days to ex-date
—
Last cash dividend
—
Median ex-date open gap
—
Reported payment date
—
Waiting for analysis. The ex-dividend date is the key date for the mechanical price adjustment; the payment date is when eligible holders receive the cash.
Recent ex-dividend sessions
Ex-date
Cash dividend
Dividend / prior close
Open gap
Close-to-close
Gap after dividend reference
Complete or review the 10-gate workflow, then generate the decision packet.
Why the model says this
Blockers / what changes the decision
Personal learning: no outcome calibration yet. Reviewed journal decisions will gradually build a local evidence table. This is empirical calibration, not a trained ML model yet.
Personal Learning Engine
Every completed one-click analysis is saved locally as a model observation. Later, Noweir can measure what actually happened 5D, 1M and 3M afterward and compare which signal combinations were more or less useful.
Observations
0
Independent 1M outcomes
0
Median next 1M
—
Directional hit rate
—
Calibration stage
Collecting
Collecting evidence. The system needs repeated observations and matured future outcomes before it can show meaningful personal calibration.
Personal Learning V2: waiting for evaluation. V2 adds chronological stability checks and keeps all adaptive influence at zero until at least 20 independent matured ~1-month outcome windows exist.
Adaptive Weighting
Learns which evidence combinations have been more or less useful in your matured local sample. It stays in shadow mode until enough outcomes exist, then applies only a small capped adjustment so a noisy sample cannot overpower the core model.
SHADOW MODE
Matured sample
0
Active factors
0
Direction adjustment
0.00
Confidence adjustment
0
Max influence
0%
Shadow mode: no adaptive weight is applied until at least 20 independent matured ~1-month outcome windows exist.
Current factor adjustments
What appears stronger or weaker in your local sample
Recent tracked observations
Date
Ticker
Stance
Bias
Confidence
Entry timing
1M outcome
V1.2 Decision Integrity
Separates evidence completeness from business fundamentals, trend, setup quality, planned-entry quality and current-price attractiveness. A high evidence score no longer looks like a high-probability trade.
Research state
—
Evidence coverage
—
Fundamental snapshot
—
Trend
—
Setup quality
—
Planned entry
—
Current price
—
Data quality
—
Waiting for V1.2 analysis.
Cross-layer conflicts
News quality
SEC intelligence
AI Research Packet & System Check
Builds the V1.2.1 consolidated packet with stable research layers plus decision integrity, smarter conflicts, data-quality guard, event-aware news clustering, scoped SEC intelligence, effective-sample diagnostics and semantic regression checks. Read-only MCP research tools are available in production.
Packet coverage
—
Available layers
—
Warnings
—
Smoke checks
—
Status
Ready
Ready. Build the packet on demand. The normal one-click analysis does not need this extra network call.
Advanced tools, goals, screeners, portfolio lab and journals
Goal Path Planner ?
Goal math + realism check
Current portfolio value
Target value
Months to target
Monthly contribution
The target is useful for planning, but the system will also push you toward process goals you can control.
Theme Discovery + Asymmetry Lab ?
V1.5 · discovery feeds the same decision engine
The score is a research funnel, not a probability: factor evidence + chronological 21D/63D model + theme/dependency context + portfolio correlation. This browser also builds a real point-in-time valuation history from your own runs instead of backfilling old multiples from today's data.
Optional sourced expectations / DCF inputs
Power-user inputs only. Paste JSON keyed by ticker. Expectation inputs should contain sourced actual/consensus/prior-consensus values. DCF inputs require sourced FCF forecasts, WACC and terminal growth. Missing assumptions stay missing.
Choose a theme to see its dependency chain and market-evidence stage.
Institutional workflow: theme → industry KPIs → peer comps → shortlist → deep evidence → falsifiable thesis → catalysts → portfolio risk → human review.
Ticker
Research state
Score
Coverage
Value
Quality
Growth
Momentum
ML 21D
ML 63D
Valuation context
Portfolio corr.
Why is this moving? ?
Evidence-first catalyst check
Run the analysis to get a plain-English explanation of what the numbers mean and what to investigate next.
Guided research path
1What happened?
Run the analysis to summarize the price move and volume.
2Is it unusual?
Compare the move with normal volatility and momentum.
3Stock or sector?
Compare the stock with its sector benchmark.
4Possible catalyst?
Check recent headlines and official filings.
5Verify before concluding
Cross-check independent evidence and unresolved conflicts.
Candidate news catalysts
Recent SEC filings
SEC filing monitor ?
Official EDGAR
Form
Date
Details
Advanced screener ?
Rule syntax · up to 30 tickers
Examples: rsi<55 · relvol>1.5 · price>sma200 · return1m>5 · pe<30 · or phrases like RSI under 55 and relative volume above 1.5
Match
Ticker
Price
1D ?
1M ?
RSI ?
Rel Vol ?
Failed checks ?
Watchlist workspace
This development version intentionally uses browser persistence. Cross-device cloud sync will be connected only after we add authenticated durable storage.
Market Memory ?
Historical analogs · descriptive, not predictive
What this teaches: instead of asking “what usually happens after RSI is 54?”, Market Memory asks a richer question: “when this same stock previously had a similar combination of return, volume, volatility, RSI and moving-average position, what happened next?” Historical similarity is context, not a forecast.
Past date
Similarity ?
Next 5D
Next 1M
Next 3M
Closest matching features
Choose a ticker and click “Find similar past setups.”
Evidence Graph + Conflict Lens ?
Cross-check evidence layers
Confirmations
Conflicts / missing explanations
Evidence Delta ?
What changed since your last check?
Baseline snapshots stay in this browser. This intentionally avoids pretending we have cloud history before authenticated storage is connected.
Watchlist Impact Radar ?
Map recent world news to your tracked stocks
Research Attention Queue ?
What deserves attention first?
Score
Ticker
1D
Rel Vol
Why
Watchlist Event Radar ?
Earnings & event metadata
Ticker
Event
When
Days
Event dates are Yahoo-derived and should be verified with issuer or exchange sources before consequential decisions.
Portfolio Decision Packet ?
Recommended portfolio workflow
One portfolio in one base currency, one trigger. Run portfolio now orchestrates concentration, ETF look-through, themes, cash rules, entry intelligence and catalyst checks for the full portfolio, then automatically runs the deeper V1.2.1 stack (news quality, SEC, Market Memory, historical backtest, conflicts and data quality) on candidates and the highest-attention holdings. It does not place trades and it never invents halal status.
Base currency
Cash
Cash floor %
Max position %
Holdings
Preferred: TICKER | SHARES | COST/SHARE | CURRENCY | HALAL STATUS | REALIZED P/L | ROLE. Quick input such as MU 1 share is also accepted, but missing cost basis/currency/halal status remains unknown rather than being invented.
New candidates to test (optional)
One line: TICKER | HALAL STATUS | CURRENCY | ROLE. Candidates have zero current weight and are tested for BUY REVIEW vs WATCH / NO BUY.
Optional executed-trade journal
One line: DATE | TICKER | ACTION | SHARES | PRICE | CURRENCY | REALIZED P/L | REASON
Run portfolio to calculate cash action and the highest-gate-clearing new setup, if one exists.
Portfolio risk decomposition
Risk contributors & correlated clusters
Model disagreement
Historical covariance, volatility, drawdown and Expected Shortfall are descriptive risk estimates, not guaranteed future loss bounds. Proxy factor betas are not MSCI Barra-equivalent attribution.
Holding decisions
Ticker
Research state
Weight
Unrealized
Halal
GTS
Current R/R
Catalyst
Portfolio flags
Theme exposure
Market data: the current provider attempts Yahoo 1-minute intraday bars when available, but it is not a guaranteed exchange-grade live feed. Exact quote age is included in every holding packet. Privacy: Save locally uses this browser only; the portfolio endpoint processes the POST payload for the request and does not intentionally persist account data server-side.
Portfolio Concentration Detail ?
Portfolio-level risk context
Why this exists: owning many tickers does not automatically mean you are diversified. This view checks name concentration, sector concentration, HHI, event collisions and where current news is clustering across the portfolio.
Positions and weights
One line per position: TICKER WEIGHT. Weights can be percentages or relative numbers; the system normalizes them to 100%.
Portfolio observations
Positions
Ticker
Weight
Sector
1D
Attention
News
Sector exposure
Sector
Weight
Upcoming earnings / event clustering
Load your watchlist equally or enter actual weights, then click Analyze portfolio.
ETF Look-Through + Theme Collision Radar ?
Hidden exposure detector
Why this matters: owning an ETF and the same company directly can make your real exposure larger than it looks. This tool estimates direct + indirect overlap and shows where multiple holdings collide around the same sector or theme.
Hidden direct + ETF overlap
Ticker
Direct
Indirect
Effective
Effective sector exposure
Sector
Effective weight
Visible theme exposure
Collision flags
Run Portfolio Command Center first, then click Analyze hidden exposure.
Portfolio Shock Simulator ?
Scenario planning · not prediction
What this teaches: instead of asking “what will happen?”, define a scenario and see where your portfolio is vulnerable. Example: if semiconductors fell 10%, which holdings would drive the damage? These are assumptions you choose, not predicted outcomes.
Portfolio-wide shock % ?
Ticker shocks
Format: TICKER:SHOCK. Ticker-specific assumptions override all other scenario layers.
Sector shocks
Theme shocks
Ticker
Weight
Applied shock
Why
Portfolio contribution
Shock / ATR
Choose or edit a scenario, then click Run stress test. The system uses ticker > theme > sector > market precedence so one position is not shocked twice.
Decision Journal + Thesis Tracker ?
Process memory · local browser storage
Why this matters: a profitable trade can come from bad reasoning, and a losing trade can come from a disciplined process. This journal separates decision quality from outcome. This journal checklist is separate from the automated 10-gate research workflow above; it records your own thesis and post-decision process.
My thesis ?
Invalidation ?
Catalysts / event risk
Red-team: strongest case I am wrong
10-gate decision checklist ?
Framework readiness: complete the checklist. This is a process check, not a buy/sell signal.
You do not need to remember to capture evidence first. Saving now automatically stores the current evidence baseline so future Thesis Drift comparisons have a valid starting point.
Saved decisions
Date
Ticker
Intent
Process
Thesis
Drift
Outcome
Review outcome
Decision Quality Coach: save the outcome to compare process quality with result quality and guard against outcome bias.
Thesis Drift Monitor
Compares the evidence captured when you saved the decision with the market evidence now. It does not automatically decide whether your written thesis is still valid.
Original thesis
Original invalidation
Objective evidence changes
My updated judgment
Workspace Backup
Protect layouts, alerts and baselines
Exports your local watchlist, smart alerts, current symbol and saved Evidence Delta baselines so a redesign or browser reset does not silently wipe your workflow.
Alert Audit Trail
Did the alert actually fire?
Smart Alert Center ?
Browser polling · 60 second checks while this page is open