Financing (Python utilities)

A small “entry-point” repo that hosts my Python finance utilities and shared helpers used across my other projects.

Right now, the main utility here is Swing Tickers (Universe Builder): it takes a raw universe of Canadian tickers and filters/ranks the ones that look suitable for 1–3 week swing trading, using daily data from Yahoo Finance via yfinance.

Disclaimer: This is not financial advice. It’s a screening/ranking tool, not a trading system.

---

Repo layout

.
├─ data/
│  └─ can_tickers                 # one ticker per line (Yahoo format: .TO/.V/.CN/...)
├─ py/
│  └─ swing_tickers.py            # universe builder / screener
│  └─ tickers_info.py             # ticker metadata/market info builder
├─ requirements.txt
└─ README.md

---

Requirements

• Python 3.10+ recommended
• Packages:
  • yfinance
  • pandas
  • numpy

Install:

python -m venv .venv
source .venv/bin/activate   # (Linux/macOS)

pip install -r requirements.txt

---

Swing Tickers (Universe Builder)

What it does

Given a text file of tickers (one per line), it downloads daily OHLCV history and computes a set of liquidity + volatility + trend/RS checks. It then outputs:

1. Tradable tickers (newline-separated) — good for feeding into other scanners/tools
2. Tradable tickers (comma-separated) — one-line format (handy for quick copy/paste)
3. Rejected tickers CSV — includes rejection reasons for diagnostics

Input format

• One ticker per line
• Yahoo symbols (for Canada commonly: .TO, .V, .CN, sometimes .NE, etc.)
• Example:

BCE.TO
ENB.TO
CVE.TO
SHOP.TO

Repo already includes a universe file at:

• data/can_tickers

---

How to run

Option A — run as-is

Run it from root.

python swing_tickers.py

It will write to:

• out/can_tickers_swing (newline-separated tickers)
• out/can_tickers_swing_one_line (comma-separated tickers)
• out/can_tickers_rejected.csv (rejected tickers + reasons)

It will also print:

• counts (tradable/rejected/total)
• Top 20 tradable with key metrics
• a rejection reason breakdown

Option B — run from code (IDE / other scripts)

from swing_tickers import UniverseBuilderConfig, Thresholds, run_universe_builder

cfg = UniverseBuilderConfig(
    tickers_path="../data/can_tickers",
    benchmark="XIU.TO",
    out_file_path="../out/can_tickers_swing",
    out_one_line_file_path="../out/can_tickers_swing_one_line",
    out_rejected_file_path="../out/can_tickers_rejected.csv",
    period="1y",
    interval="1d",
    auto_adjust=True,
    batch_size=80,
    sleep_seconds=1.0,
    thresholds=Thresholds(
        min_price=1.0,
        min_avg_dollar_vol_20=1_000_000.0,
        max_atr_pct_14=0.05,
        max_one_day_drop_126=-0.15,
        require_above_50d=True,
        prefer_above_200d=True,
        max_stale_days=5,
    ),
)

df_tradable, df_rejected = run_universe_builder(cfg)
print(df_tradable.head())

---

Tickers Info Builder

What it does

Given a text file of tickers (one per line), it fetches metadata from Yahoo Finance and writes a CSV with:

• ticker
• exchange
• company_name
• aliases (ticker without exchange suffix, e.g. RY.TO -> RY)
• sector
• average_volume
• market_cap
• last_price
• spread_estimate

spread_estimate is computed as ask - bid when both are available; otherwise it falls back to fallback_spread_pct * last_price (default: 1%).

How to run

Run it from repo root:

python py/tickers_info.py --input data/can_tickers --out out/can_tickers_info.csv

Optional arguments:

• --fallback-spread-pct (default 0.01)
• --batch-size (default 80)
• --sleep (default 1.0 seconds)

Output

• out/can_tickers_info.csv
  • one row per ticker with metadata + liquidity-related fields

---

Filtering & scoring logic (high-level)

The screener is designed to find names that are:

• liquid enough to trade (proxy: price × volume)
• not too volatile for practical swing stops (ATR% constraint)
• not obviously “falling off a cliff” (worst 1-day drop over ~6 months)
• preferably aligned with trend (above key moving averages)
• showing some relative strength vs a benchmark (default: XIU.TO)
• avoiding stale/illiquid tickers (last bar not too old)

Key thresholds (defaults in Thresholds):

• min_price
  Minimum last close (default 1.0)

• min_avg_dollar_vol_20
  20-day average dollar volume proxy (default 1,000,000)

• max_atr_pct_14
  14-day ATR normalized by price (default 0.05 = 5%)

• max_one_day_drop_126
  Worst daily return over ~126 trading days (default -0.15)

• require_above_50d
  Hard gate: require above SMA50 (default True)

• prefer_above_200d
  Soft preference: above SMA200 boosts score (default True)

• max_stale_days
  Reject if last bar is older than N trading days (default 5)

Also included:

• Volume trend check (vol_sma20 > vol_sma50)
• Relative strength vs benchmark (rs_1m, rs_3m)
• SMA50 slope normalized by mean price (cross-ticker comparable)

---

Output files

After a run, you’ll have:

• out/can_tickers_swing

  • one ticker per line (easy to load)

• out/can_tickers_swing_one_line

  • single line, comma-separated tickers

• out/can_tickers_rejected.csv

  • rejected symbols + one or more reject_reasons

Tip: when tuning thresholds, the rejected CSV is the fastest way to see why names are failing.

---

Notes on data quality

• The script uses auto_adjust=True by default to reduce issues from splits/dividends when computing SMA/ATR/returns.
• Yahoo data can be missing or inconsistent for some tickers; the code guards against common failure modes and tracks rejections.

---

Related projects

This repo is intended to be a shared hub for finance-related tooling.

• Stage Radar: https://github.com/ChernyshovYuriy/stage-radar

• Point & Figure System: https://github.com/ChernyshovYuriy/pfsystem

• TSX Canadian Stock Screener: https://github.com/ChernyshovYuriy/stock-scanner

---

License

• MIT