ismailtasdelen/historical-gold-prices
Historical Gold Price Dataset A comprehensive, research-grade collection of gold price observations spanning from 1833 to 2026 — over 193 years of continuous data. Dataset Summary This dataset provides clean, structured, machine-readable historical gold data suitable for: Machine learning and time-series forecasting Quantitative finance research Financial data analysis Economic research Gold price prediction Correlation analysis Financial education Algorithmic… See the full description on the dataset page: https://huggingface.co/datasets/ismailtasdelen/historical-gold-prices.
Historical Gold Price Dataset
A comprehensive, research-grade collection of gold price observations spanning from 1833 to 2026 — over 193 years of continuous data.
Dataset Summary
This dataset provides clean, structured, machine-readable historical gold data suitable for:
- Machine learning and time-series forecasting
- Quantitative finance research
- Financial data analysis
- Economic research
- Gold price prediction
- Correlation analysis
- Financial education
- Algorithmic trading research
- LLM and AI training
Dataset Structure
Data Files
Key Columns (gold_prices.csv)
- date: Observation date (YYYY-MM-DD)
- gold_price_usd: Gold price in USD per troy ounce
- gold_price_usd_per_gram: Gold price in USD per gram
- gold_price_usd_per_kg: Gold price in USD per kilogram
- gold_price_indexed: Price indexed to 100 at start of series
- daily_return: Daily/monthly percentage return
- log_return: Logarithmic return
- volatility_30d/90d/1y: Annualized rolling volatility
- ma_7d/30d/50d/100d/200d: Moving averages
- rsi_14d: 14-period Relative Strength Index
- macd/macd_signal/macd_histogram: MACD indicators
- drawdown: Drawdown from all-time high
- momentum_7d/14d/30d/90d: Price momentum
Data Sources
Usage
Loading with pandas
import pandas as pd
gold_prices = pd.read_csv("gold_prices.csv", parse_dates=["date"])
macro = pd.read_csv("gold_macro.csv", parse_dates=["date"])
events = pd.read_csv("gold_events.csv", parse_dates=["event_start_date", "event_end_date"])Loading with Hugging Face datasets
from datasets import load_dataset
dataset = load_dataset("your-username/historical-gold-prices")
train = dataset["train"]Citation
@dataset{historical_gold_prices_2026,
title = {Historical Gold Price Dataset},
author = {Dataset Engineering Pipeline},
year = {2026},
publisher = {Hugging Face},
note = {Gold prices from 1833 to 2026 with macroeconomic indicators}
}License
This dataset is released under the Open Data Commons Public Domain Dedication and License (PDDL).
