80 lines
2.0 KiB
Python
80 lines
2.0 KiB
Python
|
"""
|
||
|
|
Common utilities for quantitative investment strategies.
|
||
|
|
"""
|
||
|
|
import pandas as pd
|
||
|
|
from db.common import DBManager
|
||
|
|
|
||
|
|
|
||
|
|
def get_sector_data():
|
||
|
|
"""Get sector data for all stocks."""
|
||
|
|
db = DBManager()
|
||
|
|
return db.get_sector_list()
|
||
|
|
|
||
|
|
|
||
|
|
def get_stock_data():
|
||
|
|
"""Get stock data for all available stocks."""
|
||
|
|
db = DBManager()
|
||
|
|
return db.get_ticker_list()
|
||
|
|
|
||
|
|
|
||
|
|
def get_financial_data():
|
||
|
|
"""Get financial statement data."""
|
||
|
|
db = DBManager()
|
||
|
|
return db.get_fs_list()
|
||
|
|
|
||
|
|
|
||
|
|
def get_expanded_financial_data():
|
||
|
|
"""Get expanded financial statement data."""
|
||
|
|
db = DBManager()
|
||
|
|
return db.get_expanded_fs_list()
|
||
|
|
|
||
|
|
|
||
|
|
def get_price_data(interval_months=60):
|
||
|
|
"""Get price data for the specified interval."""
|
||
|
|
db = DBManager()
|
||
|
|
return db.get_price_list(interval_months)
|
||
|
|
|
||
|
|
|
||
|
|
def get_value_metrics():
|
||
|
|
"""Get value metrics for all stocks."""
|
||
|
|
db = DBManager()
|
||
|
|
return db.get_value_list()
|
||
|
|
|
||
|
|
|
||
|
|
def calculate_zscore(series):
|
||
|
|
"""Calculate z-score for a pandas series."""
|
||
|
|
return (series - series.mean()) / series.std()
|
||
|
|
|
||
|
|
|
||
|
|
def rank_by_metric(df, metric, ascending=True):
|
||
|
|
"""Rank stocks by a metric."""
|
||
|
|
return df.sort_values(by=metric, ascending=ascending)
|
||
|
|
|
||
|
|
|
||
|
|
def rank_combined(df, metrics, weights=None, ascending=True):
|
||
|
|
"""Rank stocks by combined metrics.
|
||
|
|
|
||
|
|
Args:
|
||
|
|
df: DataFrame with stock data
|
||
|
|
metrics: List of column names to use for ranking
|
||
|
|
weights: Optional weights for each metric
|
||
|
|
ascending: Direction for ranking (True = smaller is better)
|
||
|
|
|
||
|
|
Returns:
|
||
|
|
DataFrame with combined rank
|
||
|
|
"""
|
||
|
|
if weights is None:
|
||
|
|
weights = [1] * len(metrics)
|
||
|
|
|
||
|
|
# Calculate rank for each metric
|
||
|
|
ranks = pd.DataFrame(index=df.index)
|
||
|
|
for i, metric in enumerate(metrics):
|
||
|
|
ranks[f'rank_{metric}'] = df[metric].rank(ascending=ascending) * weights[i]
|
||
|
|
|
||
|
|
# Calculate combined rank
|
||
|
|
ranks['combined_rank'] = ranks.mean(axis=1)
|
||
|
|
|
||
|
|
# Combine with original data
|
||
|
|
result = pd.concat([df, ranks['combined_rank']], axis=1)
|
||
|
|
|
||
|
|
return result.sort_values('combined_rank')
|