In [5]:
1 # Importing necessary libraries
2 import yfinance as yf
3 import matplotlib.pyplot as plt
4 from xgboost import XGBRegressor
5 from sklearn.metrics import mean_squared_error, mean_absolute_err
6 import numpy as np
7
8 # Fetching data from Yahoo Finance for Microsoft (MSFT)
9 msft = yf.download("MSFT", start="2020-01-01", end="2023-12-31")
10
11 # Displaying the first few rows of the data
12 print(msft.head())
13
14 # Displaying the last few rows of the data
15 print(msft.tail())
16
17 # Length of the dataset
18 len(msft)
19
20 # Showing the data visually
21 msft['Close'].plot()
22
23 # Splitting the data into training and testing
24 train_data = msft.iloc[:int(.99*len(msft)), :]
25 test_data = msft.iloc[int(.99*len(msft)):, :]
26
27 # Defining features and target variable
28 features = ['Open', 'Volume']
29 target = 'Close'
30
31 # Training the model
32 model = XGBRegressor()
33 model.fit(train_data[features], train_data[target])
34
35 # Predicting the close price of Microsoft based on features
36 predictions = model.predict(test_data[features])
37 print('Model Predictions:', predictions)
38
39 # Displaying the actual values
40 print('Actual Values:', test_data[target])
41
42
43 # Plotting the prediction and close price
44 plt.plot(msft['Close'], label='Close Price')
45 plt.plot(np.array(test_data.index), predictions, label='Predictio
46 plt.legend()
47 plt.show()
48
49 # Calculate and print evaluation metrics
50 mse = mean_squared_error(test_data[target], predictions)
51 rmse = np.sqrt(mse)
52 mae = mean_absolute_error(test_data[target], predictions)
53
54 print(f'Mean Squared Error (MSE): {mse}')
55 print(f'Root Mean Squared Error (RMSE): {rmse}')
56 print(f'Mean Absolute Error (MAE): {mae}')
57
[*********************100%%**********************] 1 of 1 completed
Open High Low Close Adj Close \
Date
2020-01-02 158.779999 160.729996 158.330002 160.619995 154.779510
2020-01-03 158.320007 159.949997 158.059998 158.619995 152.852234
2020-01-06 157.080002 159.100006 156.509995 159.029999 153.247345
2020-01-07 159.320007 159.669998 157.320007 157.580002 151.850082
2020-01-08 158.929993 160.800003 157.949997 160.089996 154.268814
Volume
Date
2020-01-02 22622100
2020-01-03 21116200
2020-01-06 20813700
2020-01-07 21634100
2020-01-08 27746500
Open High Low Close Adj Close \
Date
2023-12-22 373.679993 375.179993 372.709991 374.579987 374.579987
2023-12-26 375.000000 376.940002 373.500000 374.660004 374.660004
2023-12-27 373.690002 375.059998 372.809998 374.070007 374.070007
2023-12-28 375.369995 376.459991 374.160004 375.279999 375.279999
2023-12-29 376.000000 377.160004 373.480011 376.040009 376.040009
Volume
Date
2023-12-22 17091100
2023-12-26 12673100
2023-12-27 14905400
2023-12-28 14327000
2023-12-29 18723000
Model Predictions: [374.31592 360.12424 375.02295 370.12485 367.31464 367.49377 367.08853
368.99023 369.79617 369.54004 373.8377 ]
Actual Values: Date
2023-12-14 365.929993
2023-12-15 370.730011
2023-12-18 372.649994
2023-12-19 373.260010
2023-12-20 370.619995
2023-12-21 373.540009
2023-12-22 374.579987
2023-12-26 374.660004
2023-12-27 374.070007
2023-12-28 375.279999
2023-12-29 376.040009
Name: Close, dtype: float64
Mean Squared Error (MSE): 35.461799762719735
Root Mean Squared Error (RMSE): 5.954981088359537
Mean Absolute Error (MAE): 5.384429931640625
In [4]:
1 import yfinance as yf
2 import matplotlib.pyplot as plt
3 import pyfolio as pf
4 import pandas as pd
5
6 # Fetching data from Yahoo Finance for Alphabet Inc. (GOOGL)
7 symbol = 'GOOGL'
8 start_date = '2023-01-01'
9 end_date = '2023-12-31'
10 googl = yf.download(symbol, start=start_date, end=end_date)
11
12 # Plotting the closing prices
13 googl['Close'].plot(figsize=(20, 5))
14 plt.title(f'{symbol} Closing Prices')
15 plt.show()
16
17 # Calculating the daily returns
18 daily_returns = googl['Close'].pct_change()
19
20 # Drop missing values
21 daily_returns.dropna(inplace=True)
22
23 # Plot the daily returns
24 plt.figure(figsize=(12, 6))
25 plt.plot(daily_returns)
26 plt.title('Daily Returns')
27 plt.xlabel('Date')
28 plt.ylabel('Percentage Returns')
29 plt.show()
30
31 # Convert daily returns to a DataFrame
32 returns_df = pd.DataFrame({'Returns': daily_returns})
33
34 # Create a simple tear sheet
35 pf.create_simple_tear_sheet(returns_df)
36
37
38
[*********************100%%**********************] 1 of 1 completed
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-4-fa903a6151cc> in <module>
33
34 # Create a simple tear sheet
---> 35 pf.create_simple_tear_sheet(returns_df)
36
37
~/opt/anaconda3/lib/python3.8/site-packages/pyfolio/plotting.py in call_w_context(*args,
50 if set_context:
51 with plotting_context(), axes_style():
---> 52 return func(*args, **kwargs)
53 else:
54 return func(*args, **kwargs)
~/opt/anaconda3/lib/python3.8/site-packages/pyfolio/tears.py in create_simple_tear_sheet
nchmark_rets, slippage, estimate_intraday, live_start_date, turnover_denom, header_rows)
357 live_start_date = ep.utils.get_utc_timestamp(live_start_date)
358
--> 359 plotting.show_perf_stats(returns,
360 benchmark_rets,
361 positions=positions,
~/opt/anaconda3/lib/python3.8/site-packages/pyfolio/plotting.py in show_perf_stats(return
ctions, turnover_denom, live_start_date, bootstrap, header_rows)
584 perf_func = timeseries.perf_stats
585
--> 586 perf_stats_all = perf_func(
587 returns,
588 factor_returns=factor_returns,
~/opt/anaconda3/lib/python3.8/site-packages/pyfolio/timeseries.py in perf_stats(returns,
ons, turnover_denom)
724 stats = pd.Series()
725 for stat_func in SIMPLE_STAT_FUNCS:
--> 726 stats[STAT_FUNC_NAMES[stat_func.__name__]] = stat_func(returns)
727
728 if positions is not None:
~/opt/anaconda3/lib/python3.8/site-packages/empyrical/stats.py in calmar_ratio(returns, p
582
583 max_dd = max_drawdown(returns=returns)
--> 584 if max_dd < 0:
585 temp = annual_return(
586 returns=returns,
~/opt/anaconda3/lib/python3.8/site-packages/pandas/core/generic.py in __nonzero__(self)
1464 @final
1465 def __nonzero__(self) -> NoReturn:
-> 1466 raise ValueError(
1467 f"The truth value of a {type(self).__name__} is ambiguous. "
1468 "Use a.empty, a.bool(), a.item(), a.any() or a.all()."
ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.
