regime-detection

Name: regime-detection
Author: agiprolabs

247installs53forks

How to install regime-detection?

agiprolabs/claude-trading-skills/regime-detection

$npx -y skills add agiprolabs/claude-trading-skills --skill regime-detection

Installs into the current project.

Summary

Market regime identification using volatility clustering, trend detection, and statistical methods for adaptive trading

Files · 1

View on GitHub

SKILL.md

1	# Regime Detection
2
3	Identify the current market regime so you can pick the right strategy, size positions correctly, and avoid deploying trend-following logic in a ranging market (or vice versa).
4
5	## Why Regime Detection Matters
6
7	Every strategy has a "home regime." A momentum strategy prints money in a clean uptrend but bleeds in a choppy range. A mean-reversion grid thrives in low-volatility consolidation but gets steamrolled by a trending breakout. Regime detection tells you which playbook to use right now.
8
9	Key benefits:
10	- Strategy selection: Route signals to the right strategy for the current environment
11	- Position sizing: Reduce exposure in hostile regimes, increase in favorable ones
12	- Stop adaptation: Wider stops in high-vol regimes, tighter in low-vol trends
13	- Drawdown control: Sit out "danger zone" regimes (high vol + no trend)
14
15	## Core Regime Dimensions
16
17	Two orthogonal axes define the four-quadrant regime model:
18
19	\| \| Low Volatility \| High Volatility \|
20	\|---\|---\|---\|
21	\| Trending \| Q1: Clean trend — best for trend following \| Q2: Volatile trend — momentum with caution \|
22	\| Ranging \| Q3: Quiet range — mean-reversion paradise \| Q4: Choppy chaos — reduce or sit out \|
23
24	A third dimension — mean-reversion tendency (Hurst exponent) — refines Q3 by telling you how reliably price reverts.
25
26	## Simple Approaches (No ML Required)
27
28	### 1. ATR Volatility Percentile
29
30	Rank the current ATR against its own recent history to get a 0–100 percentile score.
31
32	```python
33	import pandas as pd
34	import numpy as np
35
36	def atr_percentile(
37	high: pd.Series, low: pd.Series, close: pd.Series,
38	atr_period: int = 14, lookback: int = 100
39	) -> pd.Series:
40	"""ATR percentile rank over a rolling window."""
41	tr = pd.concat([
42	high - low,
43	(high - close.shift(1)).abs(),
44	(low - close.shift(1)).abs()
45	], axis=1).max(axis=1)
46	atr = tr.rolling(atr_period).mean()
47	return atr.rolling(lookback).apply(
48	lambda x: pd.Series(x).rank(pct=True).iloc[-1], raw=False
49	)
50	```
51
52	- < 25th percentile → Low volatility regime
53	- 25th–75th → Normal volatility
54	- > 75th percentile → High volatility regime
55
56	### 2. ADX Trend Strength
57
58	ADX above 25 signals a trending market; below 20 signals a range.
59
60	```python
61	def compute_adx(
62	high: pd.Series, low: pd.Series, close: pd.Series,
63	period: int = 14
64	) -> pd.Series:
65	"""Average Directional Index."""
66	plus_dm = high.diff().clip(lower=0)
67	minus_dm = (-low.diff()).clip(lower=0)
68	# Zero out when the other is larger
69	plus_dm[plus_dm < minus_dm] = 0
70	minus_dm[minus_dm < plus_dm] = 0
71
72	tr = pd.concat([
73	high - low,
74	(high - close.shift(1)).abs(),
75	(low - close.shift(1)).abs()
76	], axis=1).max(axis=1)
77
78	atr = tr.ewm(span=period, adjust=False).mean()
79	plus_di = 100 * plus_dm.ewm(span=period, adjust=False).mean() / atr
80	minus_di = 100 * minus_dm.ewm(span=period, adjust=False).mean() / atr
81	dx = 100 * (plus_di - minus_di).abs() / (plus_di + minus_di)
82	return dx.ewm(span=period, adjust=False).mean()
83	```
84
85	### 3. EMA Slope + Price Position
86
87	```python
88	def trend_direction(close: pd.Series, period: int = 20) -> pd.Series:
89	"""Returns +1 (uptrend), -1 (downtrend), 0 (neutral)."""
90	ema = close.ewm(span=period, adjust=False).mean()
91	slope = ema.diff(5) # 5-bar slope
92	above = (close > ema).astype(int)
93	direction = pd.Series(0, index=close.index)
94	direction[(slope > 0) & (above == 1)] = 1
95	direction[(slope < 0) & (above == 0)] = -1
96	return direction
97	```
98
99	### 4. Bollinger Band Width Percentile
100
101	BB width (upper - lower) / middle as a volatility proxy. A "squeeze" (low percentile) often precedes a breakout.
102
103	```python
104	def bb_width_percentile(
105	close: pd.Series, period: int = 20,
106	std_dev: float = 2.0, lookback: int = 100
107	) -> pd.Series:
108	"""Bollinger Band width percentile."""
109	sma = close.rolling(period).mean()
110	std = close.rolling(period).std()
111	width = (2 * std_dev * std) / sma
112	return width.rolling(lookback).apply(
113	lambda x: pd.Series(x).rank(pct=True).iloc[-1], raw=False
114	)
115	```
116
117	## Statistical Approaches
118
119	### Rolling Hurst Exponent
120
121	The Hurst exponent H classifies time series behavior:
122	- H < 0.4 → Mean-reverting (anti-persistent)
123	- 0.4 ≤ H ≤ 0.6 → Random walk (no exploitable structure)
124	- H > 0.6 → Trending (persistent)
125
126	Computed via the Rescaled Range (R/S) method. See `references/methodology.md` for the full derivation.
127
128	```python
129	def hurst_exponent(series: pd.Series, max_lag: int = 50) -> float:
130	"""Estimate Hurst exponent using R/S method."""
131	lags = range(2, max_lag)
132	rs_values = []
133	for lag in lags:
134	chunks = [series.iloc[i:i+lag] for i in range(0, len(series) - lag, lag)]
135	rs_list = []
136	for chunk in chunks:
137	if len

People who liked this also viewed

6 picks

Type

azure-costAzure cost management: query costs, forecast spending, optimize to reduce waste. WHEN: \"Azure costs\", \"Azure bill\", \"cost breakdown\", \"how much am I spending\", \"forecast spending\", \"optimizSkillsJul 2026275k1.3k
pricingWhen the user wants help with pricing decisions, packaging, or monetization strategy. Also use when the user mentions 'pricing,' 'pricing tiers,' 'freemium,' 'free trial,' 'packaging,' 'price increaseSkillsJul 202638k41k
stripe-integrationMulti-harness agentic plugin marketplace for Claude Code, Codex CLI, Cursor, OpenCode, GitHub Copilot, and Gemini CLISkillsJul 202611k38k
startup-financial-modelingMulti-harness agentic plugin marketplace for Claude Code, Codex CLI, Cursor, OpenCode, GitHub Copilot, and Gemini CLISkillsJul 20269.5k38k
tushare-finance专为 Claude Code 设计的 Tushare Pro 金融数据技能包，支持 220+ 个接口，用自然语言获取 A 股行情、财务数据、宏观经济等金融数据。SkillsJul 20265.0k75
azure-pricingFetches real-time Azure retail pricing using the Azure Retail Prices API (prices.azure.com) and estimates Copilot Studio agent credit consumption. Use when the user asks about the cost of any Azure seSkillsJul 20264.5k37k

regime-detection

247installs53forks

View on GitHub

How to install regime-detection?

agiprolabs/claude-trading-skills/regime-detection

$npx -y skills add agiprolabs/claude-trading-skills --skill regime-detection

Installs into the current project.

Summary

Market regime identification using volatility clustering, trend detection, and statistical methods for adaptive trading

Files · 1

View on GitHub

SKILL.md

1	# Regime Detection
2
3	Identify the current market regime so you can pick the right strategy, size positions correctly, and avoid deploying trend-following logic in a ranging market (or vice versa).
4
5	## Why Regime Detection Matters
6
7	Every strategy has a "home regime." A momentum strategy prints money in a clean uptrend but bleeds in a choppy range. A mean-reversion grid thrives in low-volatility consolidation but gets steamrolled by a trending breakout. Regime detection tells you which playbook to use right now.
8
9	Key benefits:
10	- Strategy selection: Route signals to the right strategy for the current environment
11	- Position sizing: Reduce exposure in hostile regimes, increase in favorable ones
12	- Stop adaptation: Wider stops in high-vol regimes, tighter in low-vol trends
13	- Drawdown control: Sit out "danger zone" regimes (high vol + no trend)
14
15	## Core Regime Dimensions
16
17	Two orthogonal axes define the four-quadrant regime model:
18
19	\| \| Low Volatility \| High Volatility \|
20	\|---\|---\|---\|
21	\| Trending \| Q1: Clean trend — best for trend following \| Q2: Volatile trend — momentum with caution \|
22	\| Ranging \| Q3: Quiet range — mean-reversion paradise \| Q4: Choppy chaos — reduce or sit out \|
23
24	A third dimension — mean-reversion tendency (Hurst exponent) — refines Q3 by telling you how reliably price reverts.
25
26	## Simple Approaches (No ML Required)
27
28	### 1. ATR Volatility Percentile
29
30	Rank the current ATR against its own recent history to get a 0–100 percentile score.
31
32	```python
33	import pandas as pd
34	import numpy as np
35
36	def atr_percentile(
37	high: pd.Series, low: pd.Series, close: pd.Series,
38	atr_period: int = 14, lookback: int = 100
39	) -> pd.Series:
40	"""ATR percentile rank over a rolling window."""
41	tr = pd.concat([
42	high - low,
43	(high - close.shift(1)).abs(),
44	(low - close.shift(1)).abs()
45	], axis=1).max(axis=1)
46	atr = tr.rolling(atr_period).mean()
47	return atr.rolling(lookback).apply(
48	lambda x: pd.Series(x).rank(pct=True).iloc[-1], raw=False
49	)
50	```
51
52	- < 25th percentile → Low volatility regime
53	- 25th–75th → Normal volatility
54	- > 75th percentile → High volatility regime
55
56	### 2. ADX Trend Strength
57
58	ADX above 25 signals a trending market; below 20 signals a range.
59
60	```python
61	def compute_adx(
62	high: pd.Series, low: pd.Series, close: pd.Series,
63	period: int = 14
64	) -> pd.Series:
65	"""Average Directional Index."""
66	plus_dm = high.diff().clip(lower=0)
67	minus_dm = (-low.diff()).clip(lower=0)
68	# Zero out when the other is larger
69	plus_dm[plus_dm < minus_dm] = 0
70	minus_dm[minus_dm < plus_dm] = 0
71
72	tr = pd.concat([
73	high - low,
74	(high - close.shift(1)).abs(),
75	(low - close.shift(1)).abs()
76	], axis=1).max(axis=1)
77
78	atr = tr.ewm(span=period, adjust=False).mean()
79	plus_di = 100 * plus_dm.ewm(span=period, adjust=False).mean() / atr
80	minus_di = 100 * minus_dm.ewm(span=period, adjust=False).mean() / atr
81	dx = 100 * (plus_di - minus_di).abs() / (plus_di + minus_di)
82	return dx.ewm(span=period, adjust=False).mean()
83	```
84
85	### 3. EMA Slope + Price Position
86
87	```python
88	def trend_direction(close: pd.Series, period: int = 20) -> pd.Series:
89	"""Returns +1 (uptrend), -1 (downtrend), 0 (neutral)."""
90	ema = close.ewm(span=period, adjust=False).mean()
91	slope = ema.diff(5) # 5-bar slope
92	above = (close > ema).astype(int)
93	direction = pd.Series(0, index=close.index)
94	direction[(slope > 0) & (above == 1)] = 1
95	direction[(slope < 0) & (above == 0)] = -1
96	return direction
97	```
98
99	### 4. Bollinger Band Width Percentile
100
101	BB width (upper - lower) / middle as a volatility proxy. A "squeeze" (low percentile) often precedes a breakout.
102
103	```python
104	def bb_width_percentile(
105	close: pd.Series, period: int = 20,
106	std_dev: float = 2.0, lookback: int = 100
107	) -> pd.Series:
108	"""Bollinger Band width percentile."""
109	sma = close.rolling(period).mean()
110	std = close.rolling(period).std()
111	width = (2 * std_dev * std) / sma
112	return width.rolling(lookback).apply(
113	lambda x: pd.Series(x).rank(pct=True).iloc[-1], raw=False
114	)
115	```
116
117	## Statistical Approaches
118
119	### Rolling Hurst Exponent
120
121	The Hurst exponent H classifies time series behavior:
122	- H < 0.4 → Mean-reverting (anti-persistent)
123	- 0.4 ≤ H ≤ 0.6 → Random walk (no exploitable structure)
124	- H > 0.6 → Trending (persistent)
125
126	Computed via the Rescaled Range (R/S) method. See `references/methodology.md` for the full derivation.
127
128	```python
129	def hurst_exponent(series: pd.Series, max_lag: int = 50) -> float:
130	"""Estimate Hurst exponent using R/S method."""
131	lags = range(2, max_lag)
132	rs_values = []
133	for lag in lags:
134	chunks = [series.iloc[i:i+lag] for i in range(0, len(series) - lag, lag)]
135	rs_list = []
136	for chunk in chunks:
137	if len

People who liked this also viewed

6 picks

Type