Buffett: Quality at a Fair Price
Screener Results
ROE (≥ 0.15)
D/E Ratio (≤ 0.50)
Net Margin (≥ 0.10)
Current Ratio (≥ 1.50)
P/E Ratio (≤ 20.00)
Find the companies, data, and signals worth investigating.
The data platform forinvestment professionals
Financial data, research, analytics, and model building in one environment for investment teams.
Research. Model. Invest.
Explore data, test hypotheses, build models, automate research, and monitor portfolios in one environment.
AAPL DCF Valuation
| FCFF Projection | 2025 | 2026 | 2027 | 2028 | 2029 |
|---|---|---|---|---|---|
| Revenue | 80.00 | 84.40 | 89.04 | 93.89 | 98.89 |
| EBITDA | 3.00 | 3.17 | 3.34 | 3.52 | 3.71 |
| Income Tax | 0.40 | 0.42 | 0.45 | 0.47 | 0.50 |
| NOPAT | 1.60 | 1.69 | 1.78 | 1.88 | 1.98 |
| CapEx | 0.80 | 0.84 | 0.89 | 0.94 | 0.99 |
| Free Cash Flow for Firm (FCFF) | 0.05 | 0.21 | 0.38 | 0.55 | 0.72 |
AGENTLoaded sec_insider_company_quarterly, took the last four quarters, and charted net flow.
[1]:
ticker = "AAPL"
[2]:
df = ql.load_dataset("sec_insider_company_quarterly", ticker=ticker).tail(4)[2]:
| period | net_insider_value_usd | unique_insiders_active | |
|---|---|---|---|
| 34 | Q1 2026 | -95400000 | 9 |
| 35 | Q2 2026 | 142100000 | 12 |
| 36 | Q3 2026 | -318600000 | 14 |
| 37 | Q4 2026 | -76200000 | 8 |
[3]:
# net buying against selling, by quarter df.plot.bar(x="period", y="net_insider_value_usd")
Test ideas and uncover what's driving the numbers.
Model Builder
Problem definition
Problem Type
Predict
The horizon is also the holding period.
Optimise for
Time
Date rangeFeb 2016 – Jul 2026
126 snapshots selected
Feb 2016Jul 2026
SplitDrag the dividers
Train88
Val19
Test19
Feb 2016Jan 2025Jul 2026
Metrics
Value Core
FundamentalsBV / Share
ValuationMarket Cap +14
Select algorithms
Every algorithm you pick trains as its own model on the same factors, so you can compare them.
Regression1 of 26 selected
| Algorithm | Description | Parameters | Strength | Use Case |
|---|---|---|---|---|
| AdaBoost | Fits a sequence of weak learners on repeatedly re-weighted training data; final prediction is a weighted vote. | learning_rate=0.5, n_estimators=100, loss='linear', random_state=42 | Simple and effective boosting baseline; works well with minimal tuning. | Lightweight boosting for top/bottom classification. |
| ARD Regression | Bayesian linear regression with separate precision priors for each weight; irrelevant weights are driven to zero. | compute_score=False, alpha_1=1e-06, alpha_2=1e-06, copy_X=True, fit_intercept=True, lambda_1=1e-06, lambda_2=1e-06, max_iter=300, threshold_lambda=10000, tol=0.001 | Automatic feature pruning; handles sparse selection on factor sets. | Sparse linear factor models where some features are irrelevant. |
| Bayesian Ridge | Places Gaussian priors over weights and noise variance; inferring regularization strength from data. | alpha_1=1e-06, alpha_2=1e-06, compute_score=False, copy_X=True, fit_intercept=True, lambda_1=1e-06, lambda_2=1e-06, max_iter=300, tol=0.001 | Automatically infers optimal regularization strength; no cross-validation required. | Factor model regression with automated regularization. |
Turn your research into models, workflows, and agents.
Track portfolios, exposures, and what changes over time.
Discover
Buffett: Quality at a Fair Price
Screener Results
ROE (≥ 0.15)
D/E Ratio (≤ 0.50)
Net Margin (≥ 0.10)
Current Ratio (≥ 1.50)
P/E Ratio (≤ 20.00)
Find the companies, data, and signals worth investigating.
Analyze
AAPL DCF Valuation
| FCFF Projection | 2025 | 2026 | 2027 | 2028 | 2029 |
|---|---|---|---|---|---|
| Revenue | 80.00 | 84.40 | 89.04 | 93.89 | 98.89 |
| EBITDA | 3.00 | 3.17 | 3.34 | 3.52 | 3.71 |
| Income Tax | 0.40 | 0.42 | 0.45 | 0.47 | 0.50 |
| NOPAT | 1.60 | 1.69 | 1.78 | 1.88 | 1.98 |
| CapEx | 0.80 | 0.84 | 0.89 | 0.94 | 0.99 |
| Free Cash Flow for Firm (FCFF) | 0.05 | 0.21 | 0.38 | 0.55 | 0.72 |
AGENTLoaded sec_insider_company_quarterly, took the last four quarters, and charted net flow.
[1]:
ticker = "AAPL"
[2]:
df = ql.load_dataset("sec_insider_company_quarterly", ticker=ticker).tail(4)[2]:
| period | net_insider_value_usd | unique_insiders_active | |
|---|---|---|---|
| 34 | Q1 2026 | -95400000 | 9 |
| 35 | Q2 2026 | 142100000 | 12 |
| 36 | Q3 2026 | -318600000 | 14 |
| 37 | Q4 2026 | -76200000 | 8 |
[3]:
# net buying against selling, by quarter df.plot.bar(x="period", y="net_insider_value_usd")
Test ideas and uncover what's driving the numbers.
Build
Model Builder
Problem definition
Problem Type
Predict
The horizon is also the holding period.
Optimise for
Time
Date rangeFeb 2016 – Jul 2026
126 snapshots selected
Feb 2016Jul 2026
SplitDrag the dividers
Train88
Val19
Test19
Feb 2016Jan 2025Jul 2026
Metrics
Value Core
FundamentalsBV / Share
ValuationMarket Cap +14
Select algorithms
Every algorithm you pick trains as its own model on the same factors, so you can compare them.
Regression1 of 26 selected
| Algorithm | Description | Parameters | Strength | Use Case |
|---|---|---|---|---|
| AdaBoost | Fits a sequence of weak learners on repeatedly re-weighted training data; final prediction is a weighted vote. | learning_rate=0.5, n_estimators=100, loss='linear', random_state=42 | Simple and effective boosting baseline; works well with minimal tuning. | Lightweight boosting for top/bottom classification. |
| ARD Regression | Bayesian linear regression with separate precision priors for each weight; irrelevant weights are driven to zero. | compute_score=False, alpha_1=1e-06, alpha_2=1e-06, copy_X=True, fit_intercept=True, lambda_1=1e-06, lambda_2=1e-06, max_iter=300, threshold_lambda=10000, tol=0.001 | Automatic feature pruning; handles sparse selection on factor sets. | Sparse linear factor models where some features are irrelevant. |
| Bayesian Ridge | Places Gaussian priors over weights and noise variance; inferring regularization strength from data. | alpha_1=1e-06, alpha_2=1e-06, compute_score=False, copy_X=True, fit_intercept=True, lambda_1=1e-06, lambda_2=1e-06, max_iter=300, tol=0.001 | Automatically infers optimal regularization strength; no cross-validation required. | Factor model regression with automated regularization. |
Turn your research into models, workflows, and agents.
Manage
Track portfolios, exposures, and what changes over time.
Data you can build on.
From primary-source financials to historical market data, QuantLink gives you structured, traceable data ready for research, modeling, and analysis.
NVDANASDAQ
Years Ended(USD, except EPS)FY 2026 FY 2025 FY 2024 215.9 130.5 60.9 Straight from the source
Company financials begin with the underlying filings, keeping every number connected to its source.
Data without delay
New filings and market data are processed as they’re released, so your research and models stay current.
METANASDAQ
Years Ended(except EPS)FY 2025 FY 2024 FY 2023 201.0 164.5 134.9 50.7 30.2 26.0 150.3 134.3 108.9 78.5 69.4 46.8 60.5 52.1 39.1 Consistent by design
Filings, XBRL facts, statements and financial metrics organized into a common structure.
The global market, in one view.
Global financial data, filings, and market signals connected in one continuously updating layer — so you can see what’s moving, where, and why.
- NYSE
- SEC Filings
- Momentum
- NASDAQ
- Fundamentals
- Revisions
- LSE
- Estimates
- Volatility
- Euronext
- Earnings
- Flows
- TSE
- Prices
- News
- HKEX
- Corporate Actions
- Sector Moves
- ASX
- Ownership
- Factor Exposure
- TSX
- Transcripts