def main():
#Opdracht 1
print("OPDRACHT 1")
data = pd.read_csv('data_assign_p1.csv')
titels = list(data)
#print(titels)
#print (data['obs'][0])
X = data [['obs']]
X1 = []
Y = data [['GDP_QGR']]
Z = []
for i in range (0,88):
X1.append(i)
Z.append(data['obs'][i])
plt.plot(Z, Y, linewidth=2.0)
plt.xlabel('date (1987Q2 - 2009Q1)')
plt.ylabel('GDP growth rate') #ten opzichte van jaar ervoor
plot_acf(Y, lags=4) #x = delay/tijdstip
plot_pacf(Y, lags=4)
#plot_acf(X1)
plt.show()
print("\n")
print("OPDRACHT 2")
pd.plotting.lag_plot(Y) #correlatie y(t) en y(t+1)
#decomposed = seasonal_decompose(Y, model = 'additive')
#decomposed.plot()
model = ARMA(Y, order=(1,1))
model_fit = model.fit()
print(model_fit.summary())
#data = data.join(data)
model2 = AR(Y)
model2_fit = model2.fit(3)
#print(model_fit.summary())
print('The lag value chose is: %s' % model2_fit.k_ar)
print('The coefficients of the model are:\n %s' % model2_fit.params)
print(model.loglike(model_fit.params))
ypred = model2_fit.predict(Y, start = Y.index[3])
print(ypred)
