Optimizing trading tools: ArcticDB for financial analysis

Using ArcticDB for Financial Analysis with Python and yFinance

Efficient financial data storage and analysis are essential for traders, analysts, and data scientists. ArcticDB, a high-performance database optimized for time-series data, combined with yFinance, makes it easy to fetch, store, and analyze financial data at scale.

---

Why Use ArcticDB for Financial Data?

• Fast read/write operations for large datasets

• Supports DataFrame operations like deduplication

• Versioning for historical data comparisons

• Optimized for time-series data

---

Step 1: Install Required Libraries

pip install arcticdb yfinance pandas

---

Step 2: Fetch Stock Data with yFinance

import yfinance as yf
import pandas as pd

# Fetch historical data for Apple (AAPL)
ticker = yf.Ticker("AAPL")
hist_data = ticker.history(period="5y", auto_adjust=False)
print(hist_data.head())

---

Step 3: Store Data in ArcticDB

from arcticdb import Arctic

DB_PATH = 'lmdb://./stock_db'
LIBRARY_NAME = 'financial_data'

# Connect to ArcticDB and create a library
ac = Arctic(DB_PATH)
if LIBRARY_NAME not in ac.list_libraries():
    ac.create_library(LIBRARY_NAME)

lib = ac[LIBRARY_NAME]

# Write data to ArcticDB
lib.write("AAPL", hist_data)
print("Data written to ArcticDB.")

---

Step 4: Read and Analyze Data from ArcticDB

# Read data from ArcticDB
result = lib.read("AAPL")
df = result.data
print(df.tail())

# Simple moving average (SMA) example
df['SMA_50'] = df['Close'].rolling(window=50).mean()

# Plot the stock price with SMA
df[['Close', 'SMA_50']].plot(title="AAPL Stock Price with 50-Day SMA")

---

Step 5: Update and Deduplicate Data

# New data update
new_data = ticker.history(period="1mo", auto_adjust=False)
combined = pd.concat([result.data, new_data])
filtered = combined[~combined.index.duplicated(keep='last')]

lib.write("AAPL", filtered.sort_index())
print("Data updated.")

---

Getting Started with ArcticDB

ArcticDB Tutorial in Python

ArcticDB is a high-performance database for time-series and large data storage, particularly useful for financial data like stock prices. It supports operations similar to Pandas DataFrames and offers fast read/write capabilities.

Step 1: Install ArcticDB

pip install arcticdb

---

Step 2: Setup and Connect to ArcticDB

Create a connection to ArcticDB using either LMDB (Local DB) or MongoDB.

from arcticdb import Arctic

DB_PATH = "lmdb://./my_arctic_db"  # Path for local ArcticDB
ac = Arctic(DB_PATH)  # Connect to ArcticDB
print("Available Libraries:", ac.list_libraries())

---

Step 3: Create a Library

Libraries are collections to organize different datasets.

LIBRARY_NAME = "example_library"

# Create a new library if it doesn't already exist
if LIBRARY_NAME not in ac.list_libraries():
    ac.create_library(LIBRARY_NAME)

lib = ac[LIBRARY_NAME]
print(f"Library '{LIBRARY_NAME}' created.")

---

Step 4: Write Data to ArcticDB

You can store Pandas DataFrames in ArcticDB.

import pandas as pd

# Sample DataFrame with stock prices
data = pd.DataFrame({
    'date': pd.date_range('2023-01-01', periods=5),
    'open': [100, 102, 105, 110, 120],
    'close': [102, 105, 108, 115, 125]
}).set_index('date')

symbol = "AAPL"

# Write the DataFrame to ArcticDB
lib.write(symbol, data)
print(f"Data for {symbol} written to ArcticDB.")

---

Step 5: Read Data from ArcticDB

# Check if the symbol exists
if lib.has_symbol(symbol):
    df = lib.read(symbol).data
    print(f"Data for {symbol}:\n", df)
else:
    print(f"No data found for {symbol}.")

---

Step 6: Update Data with Deduplication

# New data with some overlapping dates
new_data = pd.DataFrame({
    'date': pd.date_range('2023-01-03', periods=5),
    'open': [106, 112, 118, 123, 130],
    'close': [110, 117, 121, 130, 135]
}).set_index('date')

# Append and deduplicate
existing = lib.read(symbol).data
combined = pd.concat([existing, new_data])
filtered = combined[~combined.index.duplicated(keep='last')]

lib.write(symbol, filtered.sort_index())
print("Data updated with deduplication.")

---

Step 7: Delete Data

lib.delete(symbol)
print(f"Data for {symbol} deleted.")

---

Summary of Key Functions

Function : Description

• create_library() - Create a new library

• write() - Store DataFrame

• read() - Fetch DataFrame

• has_symbol() - Check if data exists for a symbol

• delete() - Delete data for a symbol

---

Conclusion

By integrating ArcticDB and yFinance, you can efficiently fetch, store, and analyze financial data for better decision-making. ArcticDB’s versioning and high-performance capabilities make it a game-changer for large-scale financial analysis.

Want to take your financial analysis even further? Explore more with technical indicators and machine learning techniques!