Kekkei - Advancing Financial Science For Everyone

Introduction

OHLCV (Open, High, Low, Close, Volume) is the standard format for representing price data across timeframes. Understanding OHLCV structure is essential for implementing technical indicators and strategies.

This lesson covers:

What each OHLCV component represents
How to interpret candlestick data
Timeframe aggregation and resampling
Information density in OHLCV bars
Working with OHLCV data in pandas

OHLCV Components Explained

Each bar (candlestick) in a chart contains five data points:

OHLCV Data Structure

Field	Description	Trading Significance
Open	First traded price in the period	Opening auction price discovery
High	Highest traded price in the period	Maximum buying pressure reached
Low	Lowest traded price in the period	Maximum selling pressure reached
Close	Last traded price in the period	Final consensus value; most important for analysis
Volume	Total shares/contracts traded	Participation level; confirms price moves

python

import yfinance as yf
import pandas as pd

# Download OHLCV data for Apple
aapl = yf.download('AAPL', start='2024-01-01', end='2024-01-10', progress=False)

# Display first 5 days
print("AAPL OHLCV Data (First 5 days of 2024):")
print(aapl.head())

Interpreting a Single Bar

Example: 2024-01-02

Open: $187.15 - Trading began at this price
High: $187.74 - Buyers pushed price to this maximum
Low: $184.35 - Sellers pushed price to this minimum
Close: $185.64 - Day ended here (net bearish, closed below open)
Volume: 82.3M shares - Heavy participation (above average)

Range = High - Low = $187.74 -$ 184.35 = $3.39 (1.82% range)

This shows significant intraday volatility - buyers and sellers fought for control, with sellers winning (close < open).

The Close Price: Most Important for Analysis

While all OHLCV components matter, Close is the most important for technical analysis:

Why Close Matters Most

Official settlement price: Determines end-of-day portfolio valuations
Market consensus: Represents final agreement on fair value for that period
Least manipulable: Open can be gapped by overnight news; high/low are single prints; close requires sustained pressure
Most liquid: Highest volume typically occurs near market close (closing auction)

Most indicators use Close prices: Moving averages, RSI, MACD all typically calculate using close prices.

python

# Calculate simple moving average using Close
aapl['SMA_20'] = aapl['Close'].rolling(window=20).mean()

# Compare using different price components
aapl['SMA_20_High'] = aapl['High'].rolling(window=20).mean()
aapl['SMA_20_Low'] = aapl['Low'].rolling(window=20).mean()

print("Moving averages using different OHLC components:")
print(aapl[['Close', 'SMA_20', 'SMA_20_High', 'SMA_20_Low']].tail())

Standard practice is to use Close for indicators unless you have a specific reason to use another price (e.g., using High/Low for volatility measures like ATR).

High-Low Range: Measuring Volatility

The High-Low range encapsulates intraday volatility and is crucial for volatility-based indicators.

Average True Range (ATR) Foundation

ATR (covered in Module 8) uses the high-low range to measure "true" volatility including gaps.

python

# Calculate daily range and range %
aapl['Range'] = aapl['High'] - aapl['Low']
aapl['Range_Pct'] = (aapl['Range'] / aapl['Close']) * 100

print("Daily Price Ranges:")
print(aapl[['High', 'Low', 'Range', 'Range_Pct']].head(10))

# Average range
print(f"\nAverage daily range: ${aapl['Range'].mean():.2f}")
print(f"Average range %: {aapl['Range_Pct'].mean():.2f}%")

A typical AAPL day has a 1.58% range (high to low). Days with >3% range indicate high volatility (news, earnings, market events).

Volume: The Confirmation Indicator

Volume measures participation. High volume confirms price moves; low volume suggests weak conviction.

Volume Analysis Principles

Rising price + Rising volume = Strong uptrend (buyers stepping in aggressively)
Rising price + Falling volume = Weak uptrend (may be exhaustion)
Falling price + Rising volume = Strong downtrend (sellers aggressive)
Falling price + Falling volume = Weak downtrend (may be capitulation)

python

# Calculate volume moving average to identify high/low volume days
aapl['Volume_MA'] = aapl['Volume'].rolling(window=20).mean()
aapl['Volume_Ratio'] = aapl['Volume'] / aapl['Volume_MA']

# Calculate daily returns
aapl['Return'] = aapl['Close'].pct_change() * 100

# Identify high volume days
high_volume_days = aapl[aapl['Volume_Ratio'] > 1.5].copy()

print("High Volume Days (>1.5x average):")
print(high_volume_days[['Close', 'Return', 'Volume', 'Volume_Ratio']].head(10))

Pro tip: Use volume as a confirmation filter. If your strategy signals a buy on a breakout, confirm with volume >1.5x average. Low-volume breakouts often fail (false breakouts).

Timeframe Aggregation and Resampling

OHLCV data exists at multiple timeframes: 1-minute, 5-minute, 1-hour, daily, weekly, monthly. Understanding how to aggregate is crucial.

How Aggregation Works

When converting from lower timeframe to higher:

Open: First open of the period
High: Maximum high across all bars
Low: Minimum low across all bars
Close: Last close of the period
Volume: Sum of all volumes

python

# Download daily data
daily = yf.download('AAPL', start='2023-01-01', end='2024-01-01', progress=False)

# Resample to weekly data
weekly = daily.resample('W').agg({
    'Open': 'first',
    'High': 'max',
    'Low': 'min',
    'Close': 'last',
    'Volume': 'sum'
})

print("Daily data (last 5 days):")
print(daily[['Open', 'High', 'Low', 'Close', 'Volume']].tail())

print("\nWeekly data (last 5 weeks):")
print(weekly.tail())

Notice how weekly bars aggregate the data:

Week of 2023-12-31: Open from Monday (193.61), High across entire week (194.66), Low across week (191.09), Close on Friday (192.53)
Volume: Sum of all daily volumes = 153.7M shares for the week

Information Density in Different Timeframes

Higher timeframes smooth out noise but sacrifice timeliness. Lower timeframes offer more data points but contain more noise.

Timeframe Tradeoffs

Timeframe	Bars/Year	Signal/Noise	Use Case
1-minute	~78,000	Very low	HFT, scalping (difficult)
5-minute	~15,600	Low	Day trading
1-hour	~1,950	Medium	Swing trading entry timing
Daily	~252	Medium-High	Swing trading, position trading
Weekly	~52	High	Position trading, long-term trends
Monthly	~12	Very high	Long-term regime analysis

Multi-Timeframe Analysis

Professional traders use multiple timeframes:

Higher timeframe (weekly/daily): Identify overall trend and key levels
Lower timeframe (4H/1H): Time precise entries and exits

Example: Weekly chart shows strong uptrend → use daily chart to buy pullbacks to support.

Golden rule: Your trading timeframe should be 1-2 steps below your analysis timeframe. If analyzing daily charts, trade on 4H or 1H for entries. This prevents whipsaws from noise.

Practical OHLCV Operations in Pandas

Let's see common operations you'll use throughout this course:

python

import pandas as pd
import numpy as np

# Create sample OHLCV data
data = yf.download('AAPL', start='2023-01-01', end='2024-01-01', progress=False)

# 1. Calculate typical price (average of H, L, C)
data['Typical_Price'] = (data['High'] + data['Low'] + data['Close']) / 3

# 2. Calculate true range (accounts for gaps)
data['TR'] = np.maximum(
    data['High'] - data['Low'],
    np.maximum(
        abs(data['High'] - data['Close'].shift(1)),
        abs(data['Low'] - data['Close'].shift(1))
    )
)

# 3. Body size (open-close) and wick sizes
data['Body'] = abs(data['Close'] - data['Open'])
data['Upper_Wick'] = data['High'] - np.maximum(data['Open'], data['Close'])
data['Lower_Wick'] = np.minimum(data['Open'], data['Close']) - data['Low']

# 4. Bullish/Bearish classification
data['Bullish'] = (data['Close'] > data['Open']).astype(int)

print("OHLCV Derived Features:")
print(data[['Open', 'High', 'Low', 'Close', 'Body', 'Upper_Wick', 'Lower_Wick', 'Bullish']].tail())

These derived features form the basis for many indicators and candlestick pattern recognition algorithms.

Summary

Key Takeaways

OHLCV format is standard: Open, High, Low, Close, Volume for each time period
Close is most important for indicators and analysis - represents final consensus
High-Low range measures intraday volatility
Volume confirms price moves - high volume validates trends
Timeframe aggregation combines lower bars into higher bars using first open, max high, min low, last close, sum volume
Signal-to-noise ratio increases with higher timeframes at the cost of timeliness

Next Steps

Now that you understand OHLCV structure, the next lesson covers fetching market data with yfinance: how to download real data from Yahoo Finance and prepare it for analysis in Python.

IntroductionFocusStart Focus Mode

OHLCV Components ExplainedFocusStart Focus Mode

Interpreting a Single Bar

The Close Price: Most Important for AnalysisFocusStart Focus Mode

Why Close Matters Most

High-Low Range: Measuring VolatilityFocusStart Focus Mode

Average True Range (ATR) Foundation

Volume: The Confirmation IndicatorFocusStart Focus Mode

Volume Analysis Principles

Timeframe Aggregation and ResamplingFocusStart Focus Mode

How Aggregation Works

Information Density in Different TimeframesFocusStart Focus Mode

Multi-Timeframe Analysis

Practical OHLCV Operations in PandasFocusStart Focus Mode

SummaryFocusStart Focus Mode

Key Takeaways

Next Steps

Introduction

OHLCV Components Explained

The Close Price: Most Important for Analysis

High-Low Range: Measuring Volatility

Volume: The Confirmation Indicator

Timeframe Aggregation and Resampling

Information Density in Different Timeframes

Practical OHLCV Operations in Pandas

Summary