%matplotlib notebook
import pandas as pd
import numpy as np
import requests
from io import StringIO
import matplotlib.pyplot as plt
import statsmodels.api as sm
import seaborn as sns
from datetime import datetime

url = 'https://www.metoffice.gov.uk/hadobs/hadcrut4/data/current/time_series/HadCRUT.4.6.0.0.monthly_ns_avg.txt'
header = {
  "User-Agent": "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/50.0.2661.75 Safari/537.36",
  "X-Requested-With": "XMLHttpRequest"
}
r = requests.get(url, headers=header)

#https://www.metoffice.gov.uk/hadobs/hadcrut4/data/current/series_format.html
df = pd.read_fwf(StringIO(r.text), header=None, names=['date', 'median', 'cib-', 'cib+', 'cim-', 'cim+', 'cic-', 'cic+', 'cibm-', 'cibm+', 'ci-', 'ci+'])
df['date'] = df['date'].apply(lambda x: datetime.strptime(x, '%Y/%m'))
df['month_counter'] = df.index
df.set_index('date', inplace=True)

ols_model = sm.regression.linear_model.OLS(df['median'], sm.add_constant(df['month_counter']))
r = ols_model.fit()
r.summary2()

arima_model = sm.tsa.ARIMA(df['median'], order=(1, 0, 1))
r = arima_model.fit(trend='nc')
r.summary2()

arima_model = sm.tsa.ARIMA(df['median'], order=(1, 0, 1))
r = arima_model.fit(trend='c')
r.summary2()