Volatility Indicators

Volatility indicators measure the degree of price variation over time. Higher volatility indicates greater price uncertainty and risk.

ATR - Average True Range

The Average True Range measures market volatility by averaging the true range over a specified period. True Range accounts for gaps between periods.

Formula

\[TR = \max(high - low, |high - close_{prev}|, |low - close_{prev}|)\]
\[ATR = \text{Wilder's Smoothing}(TR, period)\]

where Wilder’s Smoothing uses \(\alpha = \frac{1}{period}\).

Parameters

Parameter

Type

Default

Range

Description

period

int

14

1-100000

Number of periods for averaging

Characteristics

Property

Value

Input

OHLC (high, low, close)

Output

Single value (ATR in price units)

Lookback

period

Memory

O(period)

API Usage

from techkit import ATR

# Object-oriented
atr = ATR(period=14)
result = atr.update_ohlcv(open, high, low, close)
if result.valid:
    print(f"ATR: {result.value:.2f}")

# Batch
from techkit import talib_compat as ta
atr_values = ta.ATR(high, low, close, timeperiod=14)

const tk = require('techkit');

// Streaming
const atr = tk.atr(14);
const result = atr.updateOHLCV({open, high, low, close});

// Batch
const values = tk.ATR(high, low, close, 14);

#include <techkit/techkit.hpp>

techkit::ATR atr(14);
techkit::OHLCV bar{open, high, low, close, volume};
auto result = atr.update(bar);

tk_indicator atr = tk_atr_new(14);
tk_ohlcv bar = {open, high, low, close, volume};
tk_result r = tk_update_ohlcv(atr, &bar);
tk_free(atr);

Trading Usage

  • Stop Loss: ATR × multiplier (e.g., 2× ATR) for dynamic stop placement

  • Position Sizing: Higher ATR = larger position risk, adjust size accordingly

  • Volatility Breakouts: ATR expansion often precedes significant moves

  • Average ATR Values:

    • Stocks: 1-3% of price

    • Forex: 50-150 pips

    • Crypto: 2-5% of price

Common Periods

Period

Use Case

7

Short-term volatility

14

Standard (most common)

21

Medium-term volatility

30

Long-term volatility

NATR - Normalized Average True Range

The Normalized ATR expresses ATR as a percentage of the closing price, making it comparable across different price levels.

Formula

\[NATR = 100 \times \frac{ATR}{close}\]

Parameters

Parameter

Type

Default

Range

Description

period

int

14

1-100000

Number of periods for ATR calculation

Characteristics

Property

Value

Input

OHLC (high, low, close)

Output

Single value (percentage, e.g., 2.5 = 2.5%)

Lookback

period

Memory

O(period)

API Usage

from techkit import NATR

natr = NATR(period=14)
result = natr.update_ohlcv(open, high, low, close)
if result.valid:
    print(f"NATR: {result.value:.2f}%")

const natr = tk.natr(14);
const result = natr.updateOHLCV({open, high, low, close});

techkit::NATR natr(14);
auto result = natr.update(bar);

tk_indicator natr = tk_natr_new(14);
tk_result r = tk_update_ohlcv(natr, &bar);

Trading Usage

  • Cross-Asset Comparison: Compare volatility across different instruments

  • Relative Volatility: NATR > 3% = high volatility, < 1% = low volatility

  • Mean Reversion: High NATR often reverts to mean over time

TRANGE - True Range

The True Range is the raw volatility measure for a single period, without smoothing. It’s the building block for ATR.

Formula

\[TR = \max(high - low, |high - close_{prev}|, |low - close_{prev}|)\]

Characteristics

Property

Value

Input

OHLC (high, low, close)

Output

Single value (TR in price units)

Lookback

1 (needs previous close)

Memory

O(1)

API Usage

from techkit import TRANGE

trange = TRANGE()
result = trange.update_ohlcv(open, high, low, close)
if result.valid:
    print(f"True Range: {result.value:.2f}")

const trange = tk.trange();
const result = trange.updateOHLCV({open, high, low, close});

techkit::TRANGE trange;
auto result = trange.update(bar);

tk_indicator trange = tk_trange_new();
tk_result r = tk_update_ohlcv(trange, &bar);

Trading Usage

  • Raw Volatility: Use when you need unsmoothed volatility

  • Gap Detection: Large TR with small high-low range indicates gap

  • ATR Calculation: TRANGE is the input for ATR calculation

Comparison: ATR vs NATR vs TRANGE

Indicator

Smoothing

Normalized

Use Case

TRANGE

No

No

Raw volatility, gap detection

ATR

Yes (Wilder’s)

No

Standard volatility measure

NATR

Yes (Wilder’s)

Yes

Cross-asset comparison