樹状図を使用したクラスターの視覚化

import pandas as pd
 import requests
 
 companies={'Apple':'AAPL','Amazon':'AMZN','Facebook':'META','Tesla':'TSLA','Alphabet (Google)':'GOOGL','Shell':'SHEL','Suncor Energy':'SU',
'Exxon Mobil Corp':'XOM','Lululemon':'LULU','Walmart':'WMT','Carters':'CRI','Childrens Place':'PLCE','TJX Companies':'TJX',
'Victorias Secret':'VSCO','MACYs':'M','Wayfair':'W','Dollar Tree':'DLTR','CVS Caremark':'CVS','Walgreen':'WBA','Curaleaf':'CURLF'}

import time
 
 all_data={}
 for key,value in companies.items():
 # Replace YOUR_API_KEY with your Alpha Vantage API key
 url = f'https://www.alphavantage.co/query?function=TIME_SERIES_DAILY_ADJUSTED&symbol={value}&apikey=<YOUR_API_KEY>&outputsize=full'
 response = requests.get(url)
 data = response.json()
 time.sleep(15)
 if 'Time Series (Daily)' in data and data['Time Series (Daily)']:
 df = pd.DataFrame.from_dict(data['Time Series (Daily)'], orient='index')
 print(f'Received data for {key}')
 else:
 print("Time series data is empty or not available.")
 df.rename(columns = {'1. open':key}, inplace = True)
 all_data[key]=df[key]

# find common dates among all data frames
 common_dates = None
 for df_key, df in all_data.items():
 if common_dates is None:
 common_dates = set(df.index)
 else:
 common_dates = common_dates.intersection(df.index)
 
 common_dates = sorted(list(common_dates))
 
 # create new data frame with common dates as index
 df_combined = pd.DataFrame(index=common_dates)
 
 # reindex each data frame with common dates and concatenate horizontally
 for df_key, df in all_data.items():
 df_combined = pd.concat([df_combined, df.reindex(common_dates)], axis=1)

import numpy as np
 import scipy.cluster.hierarchy as sch
 import matplotlib.pyplot as plt
 
 # Convert correlation matrix to distance matrix
 dist_mat = 1 - df_combined.corr()
 
 # Perform top-down clustering
 clustering = sch.linkage(dist_mat, method='complete')
 cuts = sch.cut_tree(clustering, n_clusters=[3, 4])
 
 # Plot dendrogram
 plt.figure(figsize=(10, 5))
 sch.dendrogram(clustering, labels=list(df_combined.columns), leaf_rotation=90)
 plt.title('Dendrogram of Company Correlations (Top-Down Clustering)')
 plt.xlabel('Companies')
 plt.ylabel('Distance')
 plt.show()

import numpy as np
 import scipy.cluster.hierarchy as sch
 import matplotlib.pyplot as plt
 
 # Convert correlation matrix to distance matrix
 dist_mat = 1 - df_combined.corr()
 
 # Perform bottom-up clustering
 clustering = sch.linkage(dist_mat, method='ward')
 
 # Plot dendrogram
 plt.figure(figsize=(10, 5))
 sch.dendrogram(clustering, labels=list(df_combined.columns), leaf_rotation=90)
 plt.title('Dendrogram of Company Correlations (Bottom-Up Clustering)')
 plt.xlabel('Companies')
 plt.ylabel('Distance')
 plt.show()