!pip install yfinance --upgrade --no-cache-dir

!pip install PyPortfolioOpt==1.2.1

import numpy as np import matplotlib.pyplot as plt import pandas as pd import datetime as dt import yfinance as yf

symbols_list = ['SPY'] start = dt.datetime(2020,3,1) end = dt.datetime(2021,9,1) data = yf.download(symbols_list, start=start, end=end) data.info()

spy = data.reset_index() spy.info()

# filter column adjusted close spy = spy[['Date','Adj Close', 'Volume']] spy.info()

# create variables spy['day_of_week'] = spy['Date'].dt.dayofweek spy['return'] = spy['Adj Close'].pct_change() spy = spy.bfill()

from pypfopt.efficient_frontier import EfficientFrontier from pypfopt import risk_models from pypfopt import expected_returns from pypfopt import cla from pypfopt.plotting import Plotting from pypfopt.discrete_allocation import DiscreteAllocation, get_latest_prices from matplotlib.ticker import FuncFormatter import seaborn as sns from pypfopt import objective_functions

p_symbols_list = ['PYPL', 'VHT','BTC-USD', 'BILI', 'ZM', 'SPY', 'FB', 'AMZN'] #, '^IRX' p_start = dt.datetime(2020,3,1) p_end = dt.datetime(2021,9,1) p_data = yf.download(p_symbols_list, start=p_start, end=p_end)

p_df = p_data['Adj Close'] # Unique names (e.g., crypto trade on weekends as well but stock do not) p_df = p_df.ffill() p_df.head()

sharpe_portfolio_wt = {'AMZN': 0.07631, 'BILI': 0.17167, 'BTC-USD': 0.26975, 'FB': 0.0922, 'PYPL': 0.12928, 'SPY': 0.05111, 'VHT': 0.04954, 'ZM': 0.16014}

latest_prices = get_latest_prices(p_df) da = DiscreteAllocation(sharpe_portfolio_wt, latest_prices, total_portfolio_value=1000000) allocation, leftover = da.greedy_portfolio() print("Discrete allocation:", allocation) print("Funds remaining: ${:.2f}".format(leftover))

sharpe_portfolio_wt_list = list(sharpe_portfolio_wt.values()) ret_data_s = p_df.pct_change()[1:] weighted_returns_s = (sharpe_portfolio_wt_list * ret_data_s) portfolio_ret_s = pd.DataFrame(weighted_returns_s.sum(axis=1)) ret_data_s = ret_data_s.merge(portfolio_ret_s, on="Date", how="left") ret_data_s = ret_data_s.rename(columns={0: "portfolio_ret"})

ret_data_s=ret_data_s[["portfolio_ret"]]

ret_data_s = ret_data_s.reset_index() ret_data_s.info()

ret_data_s['day_of_week'] = ret_data_s['Date'].dt.dayofweek ret_data_s['return'] = ret_data_s['portfolio_ret']

ret_data_s.tail(10)

ret_data_s[ret_data_s["return"]<-100]

df = spy.merge(ret_data_s, how = "inner", on ="Date",suffixes=('_spy', '_sharpe')) df.head(10)

df['cumulative_sharpe_portfolio_ret'] = (df['return_sharpe'] + 1).cumprod() df['cumulative_spy_ret'] = (df['return_spy'] + 1).cumprod() df.tail()

sns.scatterplot('Date', 'cumulative_sharpe_portfolio_ret', data=df) sns.scatterplot('Date', 'cumulative_spy_ret', data=df) plt.legend(['cumulative_sharpe_portfolio_ret','cumulative_spy_ret'])

df['return_spy'].hist(bins=100, color='r', alpha=0.5, figsize = (16,8) ) df['return_sharpe'].hist(bins=100, color='g', alpha=0.5)

import scipy.stats as stats print("Difference in mean return: ") print((df['return_sharpe'].mean() - df['return_spy'].mean())*100) stat, p = stats.ttest_ind(df['return_sharpe'], df['return_spy'], equal_var=False, alternative='greater') # interpret p-value alpha = 0.05 print("p value is " + str(p)) if p <= alpha: print('The difference in mean return is significantly different (reject H0)') else: print('The difference in mean return is not significantly different (fail to reject H0)')