使用 SVR 预测股票价格：一个基于时间序列的问题

Question

我正在尝试使用 SVR 预测股票价格 (Adj Close)。我能够针对训练数据训练模型，但我收到测试数据错误。从 2014 年到 2018 年，训练数据存储在 dataframe df 中，从 2019 年到今天，测试数据存储在 dataframe test_df 中。这是代码：

import pandas as pd 
import pandas_datareader.data as web
import datetime
import numpy as np 
from matplotlib import style

# Get the stock data using yahoo API:
style.use('ggplot')

# get 2014-2018 data to train our model
start = datetime.datetime(2014,1,1)
end = datetime.datetime(2018,12,30)
df = web.DataReader("TSLA", 'yahoo', start, end) 

# get 2019 data to test our model on 
start = datetime.datetime(2019,1,1)
end = datetime.date.today()
test_df = web.DataReader("TSLA", 'yahoo', start, end) 


# sort by date
df = df.sort_values('Date')
test_df = test_df.sort_values('Date')

# fix the date 
df.reset_index(inplace=True)
df.set_index("Date", inplace=True)
test_df.reset_index(inplace=True)
test_df.set_index("Date", inplace=True)

df.tail()

# Converting dates

import matplotlib.dates as mdates

# change the dates into ints for training 
dates_df = df.copy()
dates_df = dates_df.reset_index()

# Store the original dates for plotting the predicitons
org_dates = dates_df['Date']

# convert to ints
dates_df['Date'] = dates_df['Date'].map(mdates.date2num)

dates_df.tail()

# Use sklearn support vector regression to predicit our data:
from sklearn.svm import SVR

dates = dates_df['Date'].to_numpy()
prices = df['Adj Close'].to_numpy()

#Convert to 1d Vector
dates = np.reshape(dates, (len(dates), 1))
prices = np.reshape(prices, (len(prices), 1))

svr_rbf = SVR(kernel= 'rbf', C= 1e3, gamma= 0.1)
svr_rbf.fit(dates, prices)


plt.figure(figsize = (12,6))
plt.plot(dates, prices, color= 'black', label= 'Data')
plt.plot(org_dates, svr_rbf.predict(dates), color= 'red', label= 'RBF model') 
plt.xlabel('Date')
plt.ylabel('Price')
plt.legend()
plt.show()

对于训练数据，它在此处工作正常。接下来，我如何预测测试数据(test_df).

Answer 1

按照您的约定，它应该如下所示：

# change the dates into ints for training 
test_dates_df = test_df.copy()
test_dates_df = test_dates_df.reset_index()

# Store the original dates for plotting the predicitons
test_org_dates = test_dates_df['Date']

# convert to ints
test_dates_df['Date'] = test_dates_df['Date'].map(mdates.date2num)

test_dates = test_dates_df['Date'].to_numpy()
test_prices = test_df['Adj Close'].to_numpy()

#Convert to 1d Vector
test_dates = np.reshape(test_dates, (len(test_dates), 1))
test_prices = np.reshape(test_prices, (len(test_prices), 1))

# Predict on unseen test data
y_hat_test = svr_rbf.predict(test_dates)

# Visualize predictions against real values
plt.figure(figsize = (12,6))
plt.plot(test_dates, test_prices, color= 'black', label= 'Data')
plt.plot(test_org_dates, y_hat_test, color= 'red', label= 'RBF model (test)') 
plt.xlabel('Date')
plt.ylabel('Price')
plt.legend()
plt.show()

使用 SVR 预测股票价格：一个基于时间序列的问题

Forecasting stock prices using SVR: A timeseries based problem

regression

time-series

svm

forecasting

stock