def main(tickers=['AAPL'], start=None, end=None, n_steps=21):
data = OrderedDict()
pred_data = OrderedDict()
forecast_data = OrderedDict()
for ticker in tickers:
data[ticker] = fc.get_time_series(ticker, start, end)
data[ticker] = fc.get_sma_regression_features(data[ticker]).dropna()
# cross-validation testing
split = rand.uniform(0.60, 0.80)
train_size = int(len(data[ticker]) * split)
train, test = data[ticker][0:train_size], data[ticker][
train_size:len(data[ticker])]
# values of features
X = list(train['sma_15'].values)
# target values
Y = list(train['adj_close'].values)
mdl = sm.OLS(Y, X).fit()
print(mdl.summary())
print(mdl.params)
print(mdl.bse)
# in sample prediction
pred = mdl.predict(test['sma_15'].values)
# summarize the fit of the model
explained_variance_score, mean_absolute_error, mean_squared_error, median_absolute_error, r2_score = fc.get_regression_metrics(
test['adj_close'].values, pred)
print("{} Ordinary Least Squares\n"
"-------------\n"
"Explained variance score: {:.3f}\n"
"Mean absolute error: {:.3f}\n"
"Mean squared error: {:.3f}\n"
"Median absolute error: {:.3f}\n"
"Coefficient of determination: {:.3f}".format(
ticker, explained_variance_score, mean_absolute_error,
mean_squared_error, median_absolute_error, r2_score))
pred_results = pd.DataFrame(data=dict(original=test['adj_close'],
prediction=pred),
index=test.index)
pred_data[ticker] = pred_results
# out-of-sample test
forecast_data[ticker] = fc.forecast_regression(model=mdl,
sample=test,
features='sma_15',
steps=n_steps)
fig = plt.figure()
ax = fig.add_subplot(111)
ax.plot(pred_data[ticker]['original'], color='red')
ax.plot(pred_data[ticker]['prediction'], color='blue')
ax.set(title='{} OLS In-Sample Prediction'.format(ticker),
xlabel='time',
ylabel='$')
ax.legend(['Original $', 'Prediction $'])
fig.tight_layout()
fig.savefig('charts/{}-OLS-In-Sample-Prediction'.format(ticker))
fig = plt.figure()
ax = fig.add_subplot(111)
ax.plot(forecast_data[ticker]['adj_close'][-n_steps:])
ax.set(title='{} Day {} OLS Out-of-Sample Forecast'.format(
n_steps, ticker),
xlabel='time',
ylabel='$')
ax.legend(tickers)
fig.tight_layout()
fig.savefig('charts/{}-Day-{}-OLS-Out-of-Sample-Forecast'.format(
n_steps, ticker))
fig = plt.figure()
ax = fig.add_subplot(111)
for ticker in tickers:
ax.plot(data[ticker]['adj_close'])
ax.set(title='Time series plot', xlabel='time', ylabel='$')
ax.legend(tickers)
fig.tight_layout()
fig.savefig('charts/stocks.png')
return forecast_data
def main(tickers=['AAPL'], start=None, end=None, n_steps=21):
"""
Main entry point of the app
"""
data = OrderedDict()
pred_data = OrderedDict()
forecast_data = OrderedDict()
for ticker in tickers:
data[ticker] = fc.get_time_series(ticker, start, end)
data[ticker] = fc.get_sma_regression_features(data[ticker]).dropna()
# cross-validation testing
split = rand.uniform(0.60, 0.80)
train_size = int(len(data[ticker]) * split)
train, test = data[ticker][0:train_size], data[ticker][train_size:len(data[ticker])]
features = ['sma_15', 'sma_50']
# values of features
X = np.array(train[features].values)
# target values
Y = list(train['adj_close'])
# fit a Naive Bayes model to the data
mdl = MLPRegressor(hidden_layer_sizes=(100, 100, 100)).fit(X, Y)
# print(mdl)
# in-sample test
pred = mdl.predict(test[features].values)
# summarize the fit of the model
explained_variance_score, mean_absolute_error, mean_squared_error, median_absolute_error, r2_score = fc.get_regression_metrics(
test['adj_close'].values, pred)
print("{} Neural Network\n"
"-------------\n"
"Explained variance score: {:.3f}\n"
"Mean absolute error: {:.3f}\n"
"Mean squared error: {:.3f}\n"
"Median absolute error: {:.3f}\n"
"Coefficient of determination: {:.3f}".format(ticker,
explained_variance_score,
mean_absolute_error,
mean_squared_error,
median_absolute_error,
r2_score))
pred_results = pd.DataFrame(data=dict(original=test['adj_close'], prediction=pred), index=test.index)
pred_data[ticker] = pred_results
# out-of-sample test
forecast_data[ticker] = fc.forecast_regression(model=mdl, sample=test, features=features, steps=n_steps)
fig = plt.figure()
ax = fig.add_subplot(111)
ax.plot(pred_data[ticker]['original'])
ax.plot(pred_data[ticker]['prediction'])
ax.set(title='{} Neural Network In-Sample Prediction'.format(ticker), xlabel='time', ylabel='$')
ax.legend(['Original $', 'Prediction $'])
fig.tight_layout()
fig.savefig('charts/{}-Neural-Network-In-Sample-Prediction.png'.format(ticker))
fig = plt.figure()
ax = fig.add_subplot(111)
ax.plot(forecast_data[ticker]['adj_close'][-n_steps:])
ax.set(title='{} Day {} Neural Network Out-of-Sample Forecast'.format(n_steps, ticker),
xlabel='time',
ylabel='$')
ax.legend(['Forecast $'])
fig.tight_layout()
fig.savefig('charts/{}-Day-{}-Neural-Network-Out-of-Sample-Forecast.png'.format(n_steps, ticker))
fig = plt.figure()
ax = fig.add_subplot(111)
for ticker in tickers:
ax.plot(data[ticker]['adj_close'])
ax.set(title='Time series plot', xlabel='time', ylabel='$')
ax.legend(tickers)
fig.tight_layout()
fig.savefig('charts/stocks.png')
return forecast_data
def main(tickers=['AAPL'], start=None, end=None, n_steps=21):
data = OrderedDict()
pred_data = OrderedDict()
forecast_data = OrderedDict()
for ticker in tickers:
data[ticker] = fc.get_time_series(ticker, start, end)
data[ticker] = fc.get_sma_regression_features(data[ticker]).dropna()
# cross-validation testing
split = rand.uniform(0.60, 0.80)
train_size = int(len(data[ticker]) * split)
train, test = data[ticker][0:train_size], data[ticker][
train_size:len(data[ticker])]
features = ['sma_15', 'sma_50']
# values of features
X = np.array(train[features].values)
# target values
Y = np.array(train['adj_close'])
mdl = DecisionTreeRegressor().fit(X, Y)
print(mdl)
'''
dot_data = export_graphviz(mdl,
out_file=None,
feature_names=list(train[features]),
class_names='outcome',
filled=True,
rounded=True,
special_characters=True)
graph = pydot.graph_from_dot_data(dot_data)
graph.write_png("charts/decision-tree-regression.png")
'''
pred = mdl.predict(test[features].values)
# summarize the fit of the model
explained_variance_score, mean_absolute_error, mean_squared_error, median_absolute_error, r2_score = fc.get_regression_metrics(
test['adj_close'].values, pred)
print("{} Decision Trees\n"
"-------------\n"
"Explained variance score: {:.3f}\n"
"Mean absolute error: {:.3f}\n"
"Mean squared error: {:.3f}\n"
"Median absolute error: {:.3f}\n"
"Coefficient of determination: {:.3f}".format(
ticker, explained_variance_score, mean_absolute_error,
mean_squared_error, median_absolute_error, r2_score))
pred_results = pd.DataFrame(data=dict(original=test['adj_close'],
prediction=pred),
index=test.index)
pred_data[ticker] = pred_results
# out-of-sample test
forecast_data[ticker] = fc.forecast_regression(model=mdl,
sample=test.copy(),
features=features,
steps=n_steps)
fig = plt.figure()
ax = fig.add_subplot(111)
ax.plot(pred_data[ticker]['original'])
ax.plot(pred_data[ticker]['prediction'])
ax.set(title='{} Decision Trees In-Sample Prediction'.format(ticker),
xlabel='time',
ylabel='$')
ax.legend(['Original $', 'Prediction $'])
fig.tight_layout()
fig.savefig(
'charts/{}-Decision-Trees-In-Sample-Prediction.png'.format(ticker))
fig = plt.figure()
ax = fig.add_subplot(111)
ax.plot(forecast_data[ticker]['adj_close'][-n_steps:])
ax.set(title='{} Day {} Decision Trees Out-of-Sample Forecast'.format(
n_steps, ticker),
xlabel='time',
ylabel='$')
ax.legend(['Forecast $'])
fig.tight_layout()
fig.savefig(
'charts/{}-Day-{}-Decision-Trees-Out-of-Sample-Forecast'.format(
n_steps, ticker))
fig = plt.figure()
ax = fig.add_subplot(111)
for ticker in tickers:
ax.plot(data[ticker]['adj_close'])
ax.set(title='Time series plot', xlabel='time', ylabel='$')
ax.legend(tickers)
fig.tight_layout()
fig.savefig('charts/stocks.png')
return forecast_data