vivek-v-rao

vivek-v-rao / OHLC-Vol

Public

Forecast volatility using OHLC volatility estimators

19
4
100% credibility
Found Apr 17, 2026 at 19 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A Python toolkit that computes various volatility measures from daily open-high-low-close stock prices, analyzes their properties, and evaluates their forecasting power using regressions and walk-forward tests.

How It Works

1
🔍 Discover the toolkit

While researching ways to better understand stock market ups and downs from daily price charts, you stumble upon this handy collection of tools shared online.

2
📦 Grab the files

Download the simple set of files, which come with ready example stock prices for familiar investments like SPY and gold.

3
📊 Compute volatility numbers

Run the easy starter tool to turn those daily open-high-low-close prices into various measures of how much prices wiggle each day.

4
📈 Spot patterns and links

Watch as it reveals summary stats, repeating patterns over time, and how different wiggle measures connect to each other.

5
🔮 Forecast future wiggles

Use the analysis tools to predict upcoming volatility days ahead, testing which measures work best in practice.

6
📋 Check real-world performance

Review out-of-sample reports showing accurate predictions, biases, and comparisons across stocks.

✅ Master volatility insights

Celebrate having clear reports on the best ways to estimate and forecast stock volatility, ready to apply to your own prices.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 19 to 19 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is OHLC-Vol?

OHLC-Vol is a Python toolkit that computes a suite of classical OHLC volatility estimators—like Parkinson, Garman-Klass, and Yang-Zhang—from daily price data, then evaluates their power to forecast future realized volatility. It solves the limitation of close-to-close returns by extracting richer intraday info for better market volatility forecasts, using non-negative least squares regressions in linear, log, sqrt-variance, and variance-space forms. Users run CLI scripts to generate OHLC volume data, build lag/MA predictors, fit models, and get in-sample R² or walk-forward OOS metrics like MAE and RMSE.

Why is it gaining traction?

It stands out with ready-to-run analysis on real assets like SPY and QQQ, including OOS forecasting benchmarks that beat simple autoregressions, plus resampling for null checks. Developers hook into CLI options for custom horizons (1/5/21 days), external predictors, or same-measure baselines, delivering quick insights on OHLC volatility formulas vs. GARCH-style dynamics. The non-negative constraints ensure realistic forecasts, with model comparisons highlighting when linear beats nonlinear.

Who should use this?

Quants prototyping volatility forecasts for options trading strategies, where implied volatility forecasts matter. Risk analysts needing OHLC vol estimators for portfolio stress tests. Traders backtesting market volatility index predictors against realized vol targets.

Verdict

Grab it for fast vol forecasting experiments—solid CLI, example data, and OOS eval make it practical despite 19 stars and 1.0% credibility signaling early maturity. Polish tests and add a forecast API for production use.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.