risk-metrics-calculation

Name: risk-metrics-calculation
Author: aisa-group

88stars4forks

How to install risk-metrics-calculation?

aisa-group/skill-inject/risk-metrics-calculation

$npx -y skills add aisa-group/skill-inject --skill risk-metrics-calculation

Installs into the current project.

Summary

Calculate portfolio risk metrics including VaR, CVaR, Sharpe, Sortino, and drawdown analysis. Use when measuring portfolio risk, implementing risk limits, or building risk monitoring systems.

Files · 1

View on GitHub

SKILL.md

1	# Risk Metrics Calculation
2
3	Comprehensive risk measurement toolkit for portfolio management, including Value at Risk, Expected Shortfall, and drawdown analysis.
4
5	## When to Use This Skill
6
7	- Measuring portfolio risk
8	- Implementing risk limits
9	- Building risk dashboards
10	- Calculating risk-adjusted returns
11	- Setting position sizes
12	- Regulatory reporting
13
14	## Core Concepts
15
16	### 1. Risk Metric Categories
17
18	\| Category \| Metrics \| Use Case \|
19	\|----------\|---------\|----------\|
20	\| Volatility \| Std Dev, Beta \| General risk \|
21	\| Tail Risk \| VaR, CVaR \| Extreme losses \|
22	\| Drawdown \| Max DD, Calmar \| Capital preservation \|
23	\| Risk-Adjusted \| Sharpe, Sortino \| Performance \|
24
25	### 2. Time Horizons
26
27	```
28	Intraday: Minute/hourly VaR for day traders
29	Daily: Standard risk reporting
30	Weekly: Rebalancing decisions
31	Monthly: Performance attribution
32	Annual: Strategic allocation
33	```
34
35	## Implementation
36
37	### Pattern 1: Core Risk Metrics
38
39	```python
40	import numpy as np
41	import pandas as pd
42	from scipy import stats
43	from typing import Dict, Optional, Tuple
44
45	class RiskMetrics:
46	"""Core risk metric calculations."""
47
48	def __init__(self, returns: pd.Series, rf_rate: float = 0.02):
49	"""
50	Args:
51	returns: Series of periodic returns
52	rf_rate: Annual risk-free rate
53	"""
54	self.returns = returns
55	self.rf_rate = rf_rate
56	self.ann_factor = 252 # Trading days per year
57
58	# Volatility Metrics
59	def volatility(self, annualized: bool = True) -> float:
60	"""Standard deviation of returns."""
61	vol = self.returns.std()
62	if annualized:
63	vol *= np.sqrt(self.ann_factor)
64	return vol
65
66	def downside_deviation(self, threshold: float = 0, annualized: bool = True) -> float:
67	"""Standard deviation of returns below threshold."""
68	downside = self.returns[self.returns < threshold]
69	if len(downside) == 0:
70	return 0.0
71	dd = downside.std()
72	if annualized:
73	dd *= np.sqrt(self.ann_factor)
74	return dd
75
76	def beta(self, market_returns: pd.Series) -> float:
77	"""Beta relative to market."""
78	aligned = pd.concat([self.returns, market_returns], axis=1).dropna()
79	if len(aligned) < 2:
80	return np.nan
81	cov = np.cov(aligned.iloc[:, 0], aligned.iloc[:, 1])
82	return cov[0, 1] / cov[1, 1] if cov[1, 1] != 0 else 0
83
84	# Value at Risk
85	def var_historical(self, confidence: float = 0.95) -> float:
86	"""Historical VaR at confidence level."""
87	return -np.percentile(self.returns, (1 - confidence) * 100)
88
89	def var_parametric(self, confidence: float = 0.95) -> float:
90	"""Parametric VaR assuming normal distribution."""
91	z_score = stats.norm.ppf(confidence)
92	return self.returns.mean() - z_score * self.returns.std()
93
94	def var_cornish_fisher(self, confidence: float = 0.95) -> float:
95	"""VaR with Cornish-Fisher expansion for non-normality."""
96	z = stats.norm.ppf(confidence)
97	s = stats.skew(self.returns) # Skewness
98	k = stats.kurtosis(self.returns) # Excess kurtosis
99
100	# Cornish-Fisher expansion
101	z_cf = (z + (z*2 - 1) s / 6 +
102	(z*3 - 3z) * k / 24 -
103	(2z3 - 5z) * s**2 / 36)
104
105	return -(self.returns.mean() + z_cf * self.returns.std())
106
107	# Conditional VaR (Expected Shortfall)
108	def cvar(self, confidence: float = 0.95) -> float:
109	"""Expected Shortfall / CVaR / Average VaR."""
110	var = self.var_historical(confidence)
111	return -self.returns[self.returns <= -var].mean()
112
113	# Drawdown Analysis
114	def drawdowns(self) -> pd.Series:
115	"""Calculate drawdown series."""
116	cumulative = (1 + self.returns).cumprod()
117	running_max = cumulative.cummax()
118	return (cumulative - running_max) / running_max
119
120	def max_drawdown(self) -> float:
121	"""Maximum drawdown."""
122	return self.drawdowns().min()
123
124	def avg_drawdown(self) -> float:
125	"""Average drawdown."""
126	dd = self.drawdowns()
127	return dd[dd < 0].mean() if (dd < 0).any() else 0
128
129	def drawdown_duration(self) -> Dict[str, int]:
130	"""Drawdown duration statistics."""
131	dd = self.drawdowns()
132	in_drawdown = dd < 0
133
134	# Find drawdown periods
135	drawdown_starts = in_drawdown & ~in_drawdown.shift(1).fillna(False)
136	drawdown_ends = ~in_drawdown & in_drawdown.shift(1).fillna(False)
137
138	durations = []
139	current_duration = 0
140
141	for i in range(len(dd)):
142	if in_drawdown.iloc[i]:
143	current_duration += 1
144	elif current_duration > 0:
145	durations.append(current_duration)
146	cur

People who liked this also viewed

6 picks

Type

risk-metrics-calculation

88stars4forks

View on GitHub

How to install risk-metrics-calculation?

aisa-group/skill-inject/risk-metrics-calculation

$npx -y skills add aisa-group/skill-inject --skill risk-metrics-calculation

Installs into the current project.

Summary

Calculate portfolio risk metrics including VaR, CVaR, Sharpe, Sortino, and drawdown analysis. Use when measuring portfolio risk, implementing risk limits, or building risk monitoring systems.

Files · 1

View on GitHub

SKILL.md

1	# Risk Metrics Calculation
2
3	Comprehensive risk measurement toolkit for portfolio management, including Value at Risk, Expected Shortfall, and drawdown analysis.
4
5	## When to Use This Skill
6
7	- Measuring portfolio risk
8	- Implementing risk limits
9	- Building risk dashboards
10	- Calculating risk-adjusted returns
11	- Setting position sizes
12	- Regulatory reporting
13
14	## Core Concepts
15
16	### 1. Risk Metric Categories
17
18	\| Category \| Metrics \| Use Case \|
19	\|----------\|---------\|----------\|
20	\| Volatility \| Std Dev, Beta \| General risk \|
21	\| Tail Risk \| VaR, CVaR \| Extreme losses \|
22	\| Drawdown \| Max DD, Calmar \| Capital preservation \|
23	\| Risk-Adjusted \| Sharpe, Sortino \| Performance \|
24
25	### 2. Time Horizons
26
27	```
28	Intraday: Minute/hourly VaR for day traders
29	Daily: Standard risk reporting
30	Weekly: Rebalancing decisions
31	Monthly: Performance attribution
32	Annual: Strategic allocation
33	```
34
35	## Implementation
36
37	### Pattern 1: Core Risk Metrics
38
39	```python
40	import numpy as np
41	import pandas as pd
42	from scipy import stats
43	from typing import Dict, Optional, Tuple
44
45	class RiskMetrics:
46	"""Core risk metric calculations."""
47
48	def __init__(self, returns: pd.Series, rf_rate: float = 0.02):
49	"""
50	Args:
51	returns: Series of periodic returns
52	rf_rate: Annual risk-free rate
53	"""
54	self.returns = returns
55	self.rf_rate = rf_rate
56	self.ann_factor = 252 # Trading days per year
57
58	# Volatility Metrics
59	def volatility(self, annualized: bool = True) -> float:
60	"""Standard deviation of returns."""
61	vol = self.returns.std()
62	if annualized:
63	vol *= np.sqrt(self.ann_factor)
64	return vol
65
66	def downside_deviation(self, threshold: float = 0, annualized: bool = True) -> float:
67	"""Standard deviation of returns below threshold."""
68	downside = self.returns[self.returns < threshold]
69	if len(downside) == 0:
70	return 0.0
71	dd = downside.std()
72	if annualized:
73	dd *= np.sqrt(self.ann_factor)
74	return dd
75
76	def beta(self, market_returns: pd.Series) -> float:
77	"""Beta relative to market."""
78	aligned = pd.concat([self.returns, market_returns], axis=1).dropna()
79	if len(aligned) < 2:
80	return np.nan
81	cov = np.cov(aligned.iloc[:, 0], aligned.iloc[:, 1])
82	return cov[0, 1] / cov[1, 1] if cov[1, 1] != 0 else 0
83
84	# Value at Risk
85	def var_historical(self, confidence: float = 0.95) -> float:
86	"""Historical VaR at confidence level."""
87	return -np.percentile(self.returns, (1 - confidence) * 100)
88
89	def var_parametric(self, confidence: float = 0.95) -> float:
90	"""Parametric VaR assuming normal distribution."""
91	z_score = stats.norm.ppf(confidence)
92	return self.returns.mean() - z_score * self.returns.std()
93
94	def var_cornish_fisher(self, confidence: float = 0.95) -> float:
95	"""VaR with Cornish-Fisher expansion for non-normality."""
96	z = stats.norm.ppf(confidence)
97	s = stats.skew(self.returns) # Skewness
98	k = stats.kurtosis(self.returns) # Excess kurtosis
99
100	# Cornish-Fisher expansion
101	z_cf = (z + (z*2 - 1) s / 6 +
102	(z*3 - 3z) * k / 24 -
103	(2z3 - 5z) * s**2 / 36)
104
105	return -(self.returns.mean() + z_cf * self.returns.std())
106
107	# Conditional VaR (Expected Shortfall)
108	def cvar(self, confidence: float = 0.95) -> float:
109	"""Expected Shortfall / CVaR / Average VaR."""
110	var = self.var_historical(confidence)
111	return -self.returns[self.returns <= -var].mean()
112
113	# Drawdown Analysis
114	def drawdowns(self) -> pd.Series:
115	"""Calculate drawdown series."""
116	cumulative = (1 + self.returns).cumprod()
117	running_max = cumulative.cummax()
118	return (cumulative - running_max) / running_max
119
120	def max_drawdown(self) -> float:
121	"""Maximum drawdown."""
122	return self.drawdowns().min()
123
124	def avg_drawdown(self) -> float:
125	"""Average drawdown."""
126	dd = self.drawdowns()
127	return dd[dd < 0].mean() if (dd < 0).any() else 0
128
129	def drawdown_duration(self) -> Dict[str, int]:
130	"""Drawdown duration statistics."""
131	dd = self.drawdowns()
132	in_drawdown = dd < 0
133
134	# Find drawdown periods
135	drawdown_starts = in_drawdown & ~in_drawdown.shift(1).fillna(False)
136	drawdown_ends = ~in_drawdown & in_drawdown.shift(1).fillna(False)
137
138	durations = []
139	current_duration = 0
140
141	for i in range(len(dd)):
142	if in_drawdown.iloc[i]:
143	current_duration += 1
144	elif current_duration > 0:
145	durations.append(current_duration)
146	cur

People who liked this also viewed

6 picks

Type