vectorbt

Name: vectorbt
Author: agiprolabs

235stars53forks

How to install vectorbt?

agiprolabs/claude-trading-skills/vectorbt

$npx -y skills add agiprolabs/claude-trading-skills --skill vectorbt

Installs into the current project.

Summary

High-performance vectorized backtesting with parameter optimization, portfolio simulation, and rich performance metrics

Files · 1

View on GitHub

SKILL.md

1	# Vectorized Backtesting with vectorbt
2
3	## Overview
4
5	vectorbt is a Python library for vectorized backtesting — running strategy simulations using NumPy/pandas array operations instead of bar-by-bar loops. This makes it 100–1000x faster than event-driven frameworks (backtrader, zipline), enabling parameter optimization across thousands of combinations in seconds.
6
7	Key strengths:
8	- Blazing speed via NumPy vectorization
9	- Built-in parameter grid search and optimization
10	- 50+ built-in performance metrics (Sharpe, Sortino, Calmar, max drawdown, profit factor)
11	- Rich plotting (equity curves, drawdowns, trade markers, heatmaps)
12	- Native pandas integration — your data stays in DataFrames throughout
13
14	## Installation
15
16	```bash
17	uv pip install vectorbt pandas numpy
18	```
19
20	vectorbt pulls in pandas, NumPy, and Plotly automatically. For technical indicators, also install pandas-ta:
21
22	```bash
23	uv pip install vectorbt pandas-ta
24	```
25
26	## Core Concepts
27
28	### 1. Signals — Boolean Entry/Exit Arrays
29
30	Strategies in vectorbt are expressed as boolean pandas Series (or arrays) indicating where to enter and exit positions:
31
32	```python
33	import vectorbt as vbt
34	import pandas as pd
35
36	# Entry: buy when fast EMA crosses above slow EMA
37	entries = fast_ema > slow_ema
38	# Exit: sell when fast EMA crosses below slow EMA
39	exits = fast_ema < slow_ema
40	```
41
42	vectorbt resolves conflicting signals automatically (you can't enter while already in a position).
43
44	### 2. Portfolio — The Backtesting Engine
45
46	`vbt.Portfolio.from_signals()` is the primary backtesting function. It takes price data and entry/exit signals, simulates trades, and computes performance:
47
48	```python
49	pf = vbt.Portfolio.from_signals(
50	close=close_prices,
51	entries=entries,
52	exits=exits,
53	init_cash=10_000,
54	fees=0.003, # 0.3% per trade
55	slippage=0.005, # 0.5% slippage
56	freq="1h", # hourly data
57	)
58	```
59
60	### 3. Metrics — Built-in Performance Analysis
61
62	```python
63	# Full stats summary
64	print(pf.stats())
65
66	# Individual metrics
67	print(f"Total Return: {pf.total_return():.2%}")
68	print(f"Sharpe Ratio: {pf.sharpe_ratio():.3f}")
69	print(f"Max Drawdown: {pf.max_drawdown():.2%}")
70	print(f"Win Rate: {pf.trades.win_rate():.2%}")
71	```
72
73	### 4. Parameter Optimization — Grid Search in Seconds
74
75	Pass arrays instead of scalars to test many parameter combos simultaneously:
76
77	```python
78	import numpy as np
79
80	fast_periods = np.arange(5, 25, 2) # 10 values
81	slow_periods = np.arange(20, 60, 5) # 8 values
82
83	fast_ma = vbt.MA.run(close, fast_periods, short_name="fast")
84	slow_ma = vbt.MA.run(close, slow_periods, short_name="slow")
85
86	# This creates 80 parameter combinations automatically
87	entries = fast_ma.ma_crossed_above(slow_ma)
88	exits = fast_ma.ma_crossed_below(slow_ma)
89	```
90
91	## Basic Workflow
92
93	### Step 1: Load OHLCV Data
94
95	```python
96	import pandas as pd
97
98	# From CSV
99	df = pd.read_csv("ohlcv.csv", parse_dates=["timestamp"], index_col="timestamp")
100	close = df["close"]
101
102	# From Yahoo Finance (traditional markets)
103	btc = vbt.YFData.download("BTC-USD", start="2023-01-01", end="2025-01-01")
104	close = btc.get("Close")
105	```
106
107	For Solana tokens, fetch data via the `birdeye-api` skill and load into a DataFrame.
108
109	### Step 2: Compute Indicators
110
111	```python
112	import pandas_ta as ta
113
114	# Using pandas-ta (see pandas-ta skill)
115	df.ta.ema(length=12, append=True)
116	df.ta.ema(length=26, append=True)
117	df.ta.rsi(length=14, append=True)
118	df.ta.bbands(length=20, std=2, append=True)
119
120	# Or using vectorbt built-ins
121	rsi = vbt.RSI.run(close, window=14)
122	bbands = vbt.BBANDS.run(close, window=20, alpha=2)
123	```
124
125	### Step 3: Generate Entry/Exit Signals
126
127	```python
128	# EMA crossover
129	entries = df["EMA_12"] > df["EMA_26"]
130	exits = df["EMA_12"] < df["EMA_26"]
131
132	# RSI mean reversion
133	entries = rsi.rsi_below(30)
134	exits = rsi.rsi_above(70)
135	```
136
137	### Step 4: Run Backtest
138
139	```python
140	pf = vbt.Portfolio.from_signals(
141	close=close,
142	entries=entries,
143	exits=exits,
144	init_cash=10_000,
145	fees=0.003,
146	slippage=0.005,
147	size=0.95, # use 95% of available cash
148	size_type="percent",
149	freq="1h",
150	)
151	```
152
153	### Step 5: Analyze Results
154
155	```python
156	# Summary statistics
157	print(pf.stats())
158
159	# Trade-level analysis
160	trades = pf.trades.records_readable
161	print(f"\nTrade count: {len(trades)}")
162	print(f"Avg holding period: {trades['Duration'].mean()}")
163
164	# Equity curve
165	pf.plot().show()
166
167	# Drawdown chart
168	pf.drawdowns.plot().show()
169	```
170
171	## Key Portfolio Parameters
172
173	\| Parameter \| Description \| Example \|
174	\|-----------\|-------------\|---------\|
175	\| `close` \| Price series (pd.Series or DataFrame) \| `df["close"]` \|
176	\| `entries` \| Boolean entry signals \| `fast > slow` \|
177	\| `exits` \| Boolean exit signals \| `fast < slow` \|
178	\| `init_cash` \| Starting capital \| `10_000` \|
179	\| `fees` \| Fee per trade (fraction) \| `0.003` (0.3%) \|
180	\| `slippage` \| Slippage per trade (fraction) \| `0.005` (0.5%) \|
181	\| `size` \| Position size \| `0.95` \|
182	\| `size_

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

vectorbt

235stars53forks

View on GitHub

How to install vectorbt?

agiprolabs/claude-trading-skills/vectorbt

$npx -y skills add agiprolabs/claude-trading-skills --skill vectorbt

Installs into the current project.

Summary

High-performance vectorized backtesting with parameter optimization, portfolio simulation, and rich performance metrics

Files · 1

View on GitHub

SKILL.md

1	# Vectorized Backtesting with vectorbt
2
3	## Overview
4
5	vectorbt is a Python library for vectorized backtesting — running strategy simulations using NumPy/pandas array operations instead of bar-by-bar loops. This makes it 100–1000x faster than event-driven frameworks (backtrader, zipline), enabling parameter optimization across thousands of combinations in seconds.
6
7	Key strengths:
8	- Blazing speed via NumPy vectorization
9	- Built-in parameter grid search and optimization
10	- 50+ built-in performance metrics (Sharpe, Sortino, Calmar, max drawdown, profit factor)
11	- Rich plotting (equity curves, drawdowns, trade markers, heatmaps)
12	- Native pandas integration — your data stays in DataFrames throughout
13
14	## Installation
15
16	```bash
17	uv pip install vectorbt pandas numpy
18	```
19
20	vectorbt pulls in pandas, NumPy, and Plotly automatically. For technical indicators, also install pandas-ta:
21
22	```bash
23	uv pip install vectorbt pandas-ta
24	```
25
26	## Core Concepts
27
28	### 1. Signals — Boolean Entry/Exit Arrays
29
30	Strategies in vectorbt are expressed as boolean pandas Series (or arrays) indicating where to enter and exit positions:
31
32	```python
33	import vectorbt as vbt
34	import pandas as pd
35
36	# Entry: buy when fast EMA crosses above slow EMA
37	entries = fast_ema > slow_ema
38	# Exit: sell when fast EMA crosses below slow EMA
39	exits = fast_ema < slow_ema
40	```
41
42	vectorbt resolves conflicting signals automatically (you can't enter while already in a position).
43
44	### 2. Portfolio — The Backtesting Engine
45
46	`vbt.Portfolio.from_signals()` is the primary backtesting function. It takes price data and entry/exit signals, simulates trades, and computes performance:
47
48	```python
49	pf = vbt.Portfolio.from_signals(
50	close=close_prices,
51	entries=entries,
52	exits=exits,
53	init_cash=10_000,
54	fees=0.003, # 0.3% per trade
55	slippage=0.005, # 0.5% slippage
56	freq="1h", # hourly data
57	)
58	```
59
60	### 3. Metrics — Built-in Performance Analysis
61
62	```python
63	# Full stats summary
64	print(pf.stats())
65
66	# Individual metrics
67	print(f"Total Return: {pf.total_return():.2%}")
68	print(f"Sharpe Ratio: {pf.sharpe_ratio():.3f}")
69	print(f"Max Drawdown: {pf.max_drawdown():.2%}")
70	print(f"Win Rate: {pf.trades.win_rate():.2%}")
71	```
72
73	### 4. Parameter Optimization — Grid Search in Seconds
74
75	Pass arrays instead of scalars to test many parameter combos simultaneously:
76
77	```python
78	import numpy as np
79
80	fast_periods = np.arange(5, 25, 2) # 10 values
81	slow_periods = np.arange(20, 60, 5) # 8 values
82
83	fast_ma = vbt.MA.run(close, fast_periods, short_name="fast")
84	slow_ma = vbt.MA.run(close, slow_periods, short_name="slow")
85
86	# This creates 80 parameter combinations automatically
87	entries = fast_ma.ma_crossed_above(slow_ma)
88	exits = fast_ma.ma_crossed_below(slow_ma)
89	```
90
91	## Basic Workflow
92
93	### Step 1: Load OHLCV Data
94
95	```python
96	import pandas as pd
97
98	# From CSV
99	df = pd.read_csv("ohlcv.csv", parse_dates=["timestamp"], index_col="timestamp")
100	close = df["close"]
101
102	# From Yahoo Finance (traditional markets)
103	btc = vbt.YFData.download("BTC-USD", start="2023-01-01", end="2025-01-01")
104	close = btc.get("Close")
105	```
106
107	For Solana tokens, fetch data via the `birdeye-api` skill and load into a DataFrame.
108
109	### Step 2: Compute Indicators
110
111	```python
112	import pandas_ta as ta
113
114	# Using pandas-ta (see pandas-ta skill)
115	df.ta.ema(length=12, append=True)
116	df.ta.ema(length=26, append=True)
117	df.ta.rsi(length=14, append=True)
118	df.ta.bbands(length=20, std=2, append=True)
119
120	# Or using vectorbt built-ins
121	rsi = vbt.RSI.run(close, window=14)
122	bbands = vbt.BBANDS.run(close, window=20, alpha=2)
123	```
124
125	### Step 3: Generate Entry/Exit Signals
126
127	```python
128	# EMA crossover
129	entries = df["EMA_12"] > df["EMA_26"]
130	exits = df["EMA_12"] < df["EMA_26"]
131
132	# RSI mean reversion
133	entries = rsi.rsi_below(30)
134	exits = rsi.rsi_above(70)
135	```
136
137	### Step 4: Run Backtest
138
139	```python
140	pf = vbt.Portfolio.from_signals(
141	close=close,
142	entries=entries,
143	exits=exits,
144	init_cash=10_000,
145	fees=0.003,
146	slippage=0.005,
147	size=0.95, # use 95% of available cash
148	size_type="percent",
149	freq="1h",
150	)
151	```
152
153	### Step 5: Analyze Results
154
155	```python
156	# Summary statistics
157	print(pf.stats())
158
159	# Trade-level analysis
160	trades = pf.trades.records_readable
161	print(f"\nTrade count: {len(trades)}")
162	print(f"Avg holding period: {trades['Duration'].mean()}")
163
164	# Equity curve
165	pf.plot().show()
166
167	# Drawdown chart
168	pf.drawdowns.plot().show()
169	```
170
171	## Key Portfolio Parameters
172
173	\| Parameter \| Description \| Example \|
174	\|-----------\|-------------\|---------\|
175	\| `close` \| Price series (pd.Series or DataFrame) \| `df["close"]` \|
176	\| `entries` \| Boolean entry signals \| `fast > slow` \|
177	\| `exits` \| Boolean exit signals \| `fast < slow` \|
178	\| `init_cash` \| Starting capital \| `10_000` \|
179	\| `fees` \| Fee per trade (fraction) \| `0.003` (0.3%) \|
180	\| `slippage` \| Slippage per trade (fraction) \| `0.005` (0.5%) \|
181	\| `size` \| Position size \| `0.95` \|
182	\| `size_

People who liked this also viewed

6 picks

Type