def run_auto_arima(self, y, **kwargs):
print(kwargs)
return auto_arima(
y,
seasonal=True,
trace=True,
error_action=
'ignore', # don't want to know if an order does not work
suppress_warnings=True, # don't want convergence warnings
stepwise=True, # set to stepwise
**kwargs)
def train_arima(X: pd.DataFrame, y: pd.Series, config: Config):
df = pd.Series(data=y.values, index=X["time_series_lag"]).sort_index()
exog = X.sort_values(
"time_series_lag")["is_working_datetime_0"].values.reshape(-1, 1)
config["model_arima"] = pm.auto_arima(
df, # exogenous=exog,
max_p=3,
max_q=3,
m=7,
seasonal=True,
error_action='ignore',
suppress_warnings=True,
)
def autoarima_method(s, pre_len, season_period):
stepwise_fit = pm.auto_arima(
np.asarray(s),
start_p=1,
start_q=1,
max_p=3,
max_q=3,
m=season_period,
start_P=0,
seasonal=True,
d=1,
D=1,
trace=True,
error_action='ignore', # don't want to know if an order does not work
suppress_warnings=True, # don't want convergence warnings
stepwise=True)
return stepwise_fit.predict(pre_len)
def autoARIMA(inputDataSet, n):
inputData = inputDataSet.astype(np.float64)
stepwise_fit = pm.auto_arima(
inputData,
start_p=1,
start_q=1,
max_p=3,
max_q=3,
m=12,
start_P=0,
seasonal=True,
d=1,
D=1,
trace=True,
error_action='ignore', # don't want to know if an order does not work
suppress_warnings=True, # don't want convergence warnings
stepwise=True) # set to stepwise
return stepwise_fit.predict(n)
def ARIMA_pred(data, forc_data):
listp = forc_data.values
stepwise_model = auto_arima(data,
start_p=1,
start_q=1,
max_p=3,
max_q=3,
start_P=0,
start_Q=0,
max_P=3,
max_Q=3,
error_action='ignore',
suppress_warnings=True,
stepwise=True)
stepwise_model.fit(data)
pr = stepwise_model.predict(n_periods=100)
plt.plot(listp, color='red', label='Real Stock Price')
plt.plot(pr, color='blue', label='Predicted Stock Price')
plt.title(' Stock Price Prediction using ARIMA')
plt.xlabel('Time')
plt.ylabel('Stock Price')
plt.legend()
plt.show()
print("For ARIMA The mean squared error is:")
print(mean_squared_error(listp, pr))
print("For ARIMA The R squared error is:")
print(r2_score(listp, pr))
plt.style.use('fivethirtyeight')
plt.scatter(data,
data - stepwise_model.predict_in_sample(data),
color="green",
s=10,
label='Train data')
plt.scatter(listp, listp - pr, color="blue", s=10, label='Test data')
plt.hlines(y=0, xmin=0, xmax=250, linewidth=2)
plt.legend(loc='upper right')
plt.title("Residual errors for ARIMA")
plt.show()
def a_arima(timeseries):
stepwise_model = auto_arima(timeseries, start_p=1, start_q=1,
max_p=3, max_q=3, m=12,
start_P=0, seasonal=True,
d=1, D=1, trace=True,
error_action='ignore',
suppress_warnings=True,
stepwise=True)
#print(stepwise_model.aic())
#df['Year'] = pd.to_datetime(df['Year'])
train = df.loc['2009-06-02 00:00:00':'2017-12-31 00:00:00']
test = df.loc['2018-01-01 00:00:00':]
# Train the Model
stepwise_model.fit(train)
future_forecast = stepwise_model.predict(n_periods=182)
future_forecast = pd.DataFrame(future_forecast, index = test.index, column=['Prediction'])
pd.concat([test,future_forecast],axis=1).iplot()
plt.xlabel('time')
plt.ylabel('passengers')
plt.plot()
train = df[:int(0.8 * (len(df)))]
test = df[int(0.8 * (len(df))):]
print(train.shape)
print(test.shape)
model = auto_arima(train,
trace=True,
start_q=0,
start_p=0,
start_Q=0,
max_p=10,
max_q=10,
max_P=10,
max_Q=10,
seasonal=True,
stepwise=False,
suppress_warnings=True,
D=1,
max_D=10,
error_action='ignore',
approximation=False)
mode.fit(train)
y_pred = model.predict(n_periods=len(valid))
from sklearn.metrics import r2_score
acc = r2_score(valid.values, y_pred)
print(acc)
result.plot()
plt.show()
# Lowess Filter also does not provide the expected results
# Using ARIMA Model to predict the time series analysis
# it is not always needed to have the HP and lowess filter analysis, but it was necessary in this case to use them
# WTI ARIMA
print(clear)
stepwise_model_wti = auto_arima(wti_train['WTI'],
start_p=1,
start_q=1,
max_p=3,
max_q=3,
m=12,
start_P=0,
seasonal=True,
d=1,
D=1,
trace=True,
error_action='ignore',
suppress_warnings=True,
stepwise=True)
print('WTI AIC')
print(stepwise_model_wti.aic())
print("Training the model on our datasets")
stepwise_model_wti.fit(wti_train['WTI'])
print("Forecasting WTI for next 13 periods")
future_forecast_wti = stepwise_model_wti.predict(n_periods=13)
dts = [
# -*- coding: utf-8 -*-
# @Time : 20/3/14 0:19
# @Author : Jay Lam
# @File : test.py
# @Software: PyCharm
import numpy as np
import pandas as pd
import warnings
warnings.simplefilter(action='ignore', category=(FutureWarning,UserWarning))
import pyramid as pm
from pyramid.datasets import load_wineind
# this is a dataset from R
wineind = pd.read_csv('pg43.csv').astype(np.float64)
#wineind =wineind = load_wineind().astype(np.float64)
# fit stepwise auto-ARIMA
stepwise_fit = pm.auto_arima(wineind, start_p=1, start_q=1,
max_p=3, max_q=3, m=12,
start_P=0, seasonal=True,
d=1, D=1, trace=True,
error_action='ignore', # don't want to know if an order does not work
suppress_warnings=True, # don't want convergence warnings
stepwise=True) # set to stepwise
#print(stepwise_fit.summary())
print(type(stepwise_fit.predict(3)))
<br/>
"""
print(__doc__)
# Author: Taylor Smith <*****@*****.**>
import pyramid as pm
import numpy as np
from matplotlib import pyplot as plt
# #############################################################################
# Load the data and split it into separate pieces
data = pm.datasets.load_wineind()
train, test = data[:150], data[150:]
# Fit a simple auto_arima model
arima = pm.auto_arima(train,
error_action='ignore',
trace=1,
seasonal=True,
m=12)
# #############################################################################
# Plot actual test vs. forecasts:
x = np.arange(test.shape[0])
plt.scatter(x, test, marker='x')
plt.plot(x, arima.predict(n_periods=test.shape[0]))
plt.title('Actual test samples vs. forecasts')
plt.show()
#divide into train and validation set
#2344 - 2000-01-01
#2487 - 2011-12-01
#2488 - 2012-01-01
#2507 - 2013-08-01
#train = dataset.loc[2344:2487:][:int(1*(len(dataset.loc[2344:2487:])))]
#test = dataset.loc[2488:2507:][int(1*(len(dataset.loc[2488:2507:]))):]
train = dataset.loc[2344:2487, :]
test = dataset.loc[2488:, :]
#-----------------------------------------------------------------------------------------------------------------
#building the model
from pyramid import auto_arima
model = auto_arima(train['Avg_Temp'],
trace=True,
error_action='ignore',
suppress_warnings=True)
#-------------------------------------------------------------------------------------------------------
forecast = model.predict(n_periods=len(test['Avg_Temp']))
forecast = pd.DataFrame(forecast,
index=test['Avg_Temp'].index,
columns=['Prediction'])
forecast.head()
#---------------------------------------------------------------------------------------------------
#PREDICT FROM 2013-Sept to 2014-Aug
forecastTest = model.predict(n_periods=24)
forecastTest = pd.DataFrame(forecastTest, columns=['PredictionUntilThisMonth'])