def test_innovations_algo_filter_kalman_filter(ar_params, ma_params, sigma2):
# Test the innovations algorithm and filter against the Kalman filter
# for exact likelihood evaluation of an ARMA process
endog = np.random.normal(size=100)
# Innovations algorithm approach
llf = arma_innovations.arma_loglike(endog, ar_params, ma_params, sigma2)
llf_obs = arma_innovations.arma_loglikeobs(endog, ar_params, ma_params,
sigma2)
score = arma_innovations.arma_score(endog, ar_params, ma_params, sigma2)
score_obs = arma_innovations.arma_scoreobs(endog, ar_params, ma_params,
sigma2)
# Kalman filter apparoach
mod = SARIMAX(endog, order=(len(ar_params), 0, len(ma_params)))
params = np.r_[ar_params, ma_params, sigma2]
# Test that the two approaches are the same
assert_allclose(llf, mod.loglike(params))
assert_allclose(llf_obs, mod.loglikeobs(params))
# Note: the tolerance on the two gets worse as more nobs are added
assert_allclose(score, mod.score(params), atol=1e-5)
assert_allclose(score_obs, mod.score_obs(params), atol=1e-5)
max_evals=50,
trials=Trials())
modelo = GBR(
n_estimators=int(best['n_estimators']),
learning_rate=best['learning_rate'],
subsample=best['x_subsample'],
alpha=best['x_alpha'],
validation_fraction=best['x_validation_fraction'],
)
modelo.fit(X_train, y_train) #MSE
y_pred = modelo.predict(X_test)
print('Mean squared error', mse(y_test, y_pred))
train_score = modelo.score(X_train, y_train) #R2
test_score = modelo.score(X_test, y_test)
print('train R2:', train_score, '-- test R2:', test_score)
pred = df_merged.drop('Stocks_price_usd',
axis=1) #Using trained model to predict data
res = modelo.predict(pred)
dffinal.drop(['Refiners_Cost_usd', 'usd', 'Stocks_price_usd', 'Pandemics'],
axis=1,
inplace=True) # 3 column df to show results
dffinal.rename(columns={'Crude_oil_price_usd': 'Real'}, inplace=True)
dffinal['Predicitions'] = res
dffinal['Difference'] = dffinal.Predicitions - dffinal.Real
preds.insert(1, "Refiners_Cost_usd", [32.72, 25.80, 29.1],
