Research through 66 lenses

One stock.Sixty-six methods.

UZI Investor Panel gives the same market and company data to 66 methodology-based roles. Value, growth, macro, technical, China-focused, momentum, quantitative, and technology lenses produce structured signals. It does not make the decision for you; it makes disagreement visible.

Explore three markets See the review process

66methodology roles

9investor schools

CN · HK · USmarket coverage

UZI://PANEL/600519.SHDEMO

Kweichow Moutai DEMO DATA

600519.SH · 64 long-book votes + 2 short-risk observers

+18.6ILLUSTRATIVE CONSENSUS

28 bullish24 neutral12 bearish

Classic value · A

+24

Growth · B

+08

Macro hedge · C

−12

China value · E

+31

A-share momentum · F

+02

Quant systematic · G

+05

Signals, scores, and comments shown here are illustrative. They are not live market data, actual runtime output, or any person’s current view.

Before / After

Put every research framework on the same table

Traditional research is not short of information. The hard part is switching, remembering, and comparing different frameworks. Investor Panel structures that disagreement.

Before · Fragmented research

Every framework in a separate place

Financials, valuation, technicals, macro, sentiment, and analyst reports live in separate places. The researcher must integrate them manually.

Financial quality

Valuation

Technical trend

Macro regime

Market sentiment

→

After · Structured review

One dataset, many methodologies

Start with one question, let each school produce a signal, then aggregate the result into a traceable distribution and consensus.

Prompt→Stock data→66 methods→Consensus

28 / 64illustrative bullish votes (long-book)

+18.6illustrative consensus score

Duan Yongping stylerepresentative lens: business quality

Macro hedgemain disagreement: cycle and valuation

Core capabilities

Capabilities supported by current code

Based on repository v3.9.4, the investor database, and evaluator structure. Counts follow current code, not legacy document headings.

01 / MULTI-METHOD

66 methodology-based roles

Each role has its own field scope, rules, and references. Outputs simulate a methodology; they are not the person’s live opinions.

02 / STRUCTURED

Structured signals

Each role can output signal, score, confidence, verdict, and comment for comparison and report assembly.

03 / SCHOOLS

Nine investor schools

Classic value, growth, macro, technical, China value, A-share momentum, quant, technology leaders, and the AI bottleneck hunter.

04 / CONSENSUS

Consensus and signal distribution

Aggregate long-book signals into bullish / neutral / bearish distributions and a consensus score, while retaining short-risk observers separately.

05 / CROSS-MARKET

Cross-market input

The workflow targets A-share, Hong Kong, and U.S. equities. Actual coverage depends on source connectivity, ticker support, and optional API configuration.

06 / TRACEABLE

Cached, traceable output

The workflow can generate raw_data, dimensions, panel, synthesis, and HTML report artifacts.

How it works

From data to panel review

Data is collected and normalized into analytical dimensions, then evaluated through role-specific rules and scoring. Outputs preserve individual differences and generate an aggregate view.

Market & Company Data

Available market, financial, valuation, industry, event, flow, and sentiment data.

Dimensions

Normalize heterogeneous data into dimensions the evaluator can consume.

Methodologies

Each role reads only fields relevant to its methodology and applies its evaluation rules.

Signals

Generate individual signal, score, confidence, verdict, and reasoning.

Panel Review

Aggregate distribution, school-level disagreement, consensus, and the full research report.

Key boundary: panel roles are research simulations built on defined methodologies and rules. Labels such as “Buffett style” or “Soros style” refer to a methodological lens, not the actual person’s present opinion on a stock.

Three market demos

Three markets, three disagreement structures

All cases below are product-interface demonstrations. They use no live market data and are not actual repository outputs. Tickers illustrate cross-market input formats only.

CASE 01 / A-SHAREILLUSTRATIVE

Quality versus valuation in a consumer leader

Scenario: a researcher wants to compare China value, classic value, macro, and momentum lenses to identify why they disagree on the same asset.

“Run the Investor Panel for 600519.SH. Show only the panel output and explain why investor schools disagree.”

$ python3 run.py 600519.SH --no-browser
# UI demo only; the command did not produce the values below

Illustrative consensus+18.6

Long-book voters64

Top disagreementValue / macro

bullish 28

43.8%

neutral 24

37.5%

bearish 12

18.8%

Duan Yongping style

China value · E

BULLISH

78 / 100

Illustrative reasoning: brand and business-model quality support a long-term view; valuation determines the margin of safety.

Zhang Mengzhu style

A-share momentum · F

BULLISH

81 / 100

Illustrative reasoning: liquidity, flow structure, and near-term catalysts matter more than long-term intrinsic value alone.

Graham style

Classic value · A

NEUTRAL

51 / 100

Illustrative reasoning: asset quality is visible, but the margin of safety is insufficient for a more positive view.

Soros style

Macro hedge · C

BEARISH

37 / 100

Illustrative reasoning: demand expectations, liquidity, and valuation feedback may amplify short-term volatility.

Input: illustrative prompt + tickerDisclosure: all numbers and comments are illustrative UI data

CASE 02 / HONG KONGILLUSTRATIVE

Platform growth versus regulatory discount

Scenario: a researcher compares growth, value, China institutional, and quantitative methods to see how long-term growth and risk discounts shape the judgment.

“Run the Investor Panel for 00700.HK and compare the core disagreements among growth, classic-value, and China-value schools.”

$ python3 run.py 00700.HK --no-browser
# UI demo only; the command did not produce the values below

Illustrative consensus+10.9

Long-book voters64

Top disagreementGrowth / quant

bullish 24

37.5%

neutral 27

42.2%

bearish 13

20.3%

Buffett style

Classic value · A

NEUTRAL

54 / 100

Illustrative reasoning: the moat is attractive, but regulation and capital allocation require a wider margin of safety.

Zhang Lei style

China value · E

BULLISH

74 / 100

Illustrative reasoning: user networks and industry position offer compounding potential; structural change matters more than short-term volatility.

Simons style

Quant systematic · G

NEUTRAL

49 / 100

Illustrative reasoning: factor signals are mixed; trend and valuation do not show a consistent statistical edge.

Cathie Wood style

Growth · B

BULLISH

69 / 100

Illustrative reasoning: platform scale and business optionality support a growth view, with an execution-risk discount.

Input: illustrative prompt + tickerDisclosure: all numbers and comments are illustrative UI data

CASE 03 / US EQUITYILLUSTRATIVE

Quality compounding versus maturity risk

Scenario: a researcher compares classic growth, technology-leader, quant, and short-risk lenses to understand why a strong brand can still attract cautious signals.

“Run the Investor Panel for AAPL and show how investor schools evaluate growth, valuation, and the technology ecosystem.”

$ python3 run.py AAPL --no-browser
# UI demo only; the command did not produce the values below

Illustrative consensus+23.8

Long-book voters64

Top disagreementTechnology / short-risk

bullish 31

48.4%

neutral 20

31.3%

bearish 13

20.3%

Fisher style

Classic value · A

BULLISH

75 / 100

Illustrative reasoning: product ecosystem, innovation culture, and customer retention support a long-term quality view.

Jensen Huang style

Technology leader · H

BULLISH

83 / 100

Illustrative reasoning: hardware-software integration and the developer ecosystem form a platform moat, while paradigm shifts deserve monitoring.

Asness style

Quant systematic · G

NEUTRAL

52 / 100

Illustrative reasoning: quality is strong, while value and momentum factors may not confirm a buy simultaneously.

Chanos style

Macro hedge · C

RISK OBSERVER

31 / 100

Illustrative reasoning: as a short-risk observer, the role watches growth maturity, crowded expectations, and valuation fragility.

Input: illustrative prompt + tickerDisclosure: all numbers and comments are illustrative UI data

Reliability & limitations

Know what it can do—and what it cannot

A financial research tool earns trust not only through rich output, but through clear boundaries around data, methods, and roles.

Methodology boundary

A research-assistance and methodology-simulation system

Role names help users understand different investment frameworks. No role output should be described as a real person’s current opinion, recommendation, or trading instruction.

All three cases on this page are interface demonstrations. Their numbers, signals, confidence values, and comments are not derived from live markets and are not investment advice.

Data sourcesThe repository integrates multiple free data sources and optional APIs. Availability depends on regional networks, rate limits, and ticker coverage.

Free fallbackSome fields can fall back to free sources, but fallback does not guarantee completeness. Missing dimensions affect score quality.

Cache structureThe workflow can cache intermediate outputs such as raw_data.json, dimensions.json, panel.json, and synthesis.json.

Data freshnessFreshness depends on upstream update schedules and cache hits. Verify timestamps and source status before use.

Roles and rulesThe 66 roles follow field allowlists, rules, and prompts. They evaluate through a methodology; they are not digital replicas of people.

Count conventionCurrent code defines 66 roles across nine groups. Two short-mandate roles act as risk observers and are excluded from the long-book consensus distribution.

Output limitsConfidence is internal confidence in an evaluation, not the probability of profit. Review output against the underlying data.

For research, not promises

Turn disagreement into a research entry point.

For individual investors, equity researchers, and research teams screening ideas, comparing frameworks, and checking blind spots.

Idea screeningEquity researchCross-framework reviewRisk discussion