errors = y.loc[start:] - y_hat MAE = np.mean(np.absolute(errors))[0] return MAE TRAIN_OFFSET = 490 TEST_OFFSET = df_train.last_valid_index() + 1 from MLP import MLP from Model import Model from var_model import VAR as VAR from FAVAR import FAVAR results = {} favar = FAVAR(factors_q) favar = FAVAR(factors_q) cols = ['CPI'] y_col = ['CPI'] # x_cols, y_col = get_cols(cols, 'GDP') X_train, y_train = df_train[cols], df_train[y_col] X_test, y_test = df_test[cols], df_test[y_col] n_factors = 1 X_train = favar.addFactors(X_train, n_factors) X_test = favar.addFactors(X_test, n_factors) lags = 4
errors = y.loc[start:] - y_hat MAE = np.mean(np.absolute(errors))[0] return MAE TRAIN_OFFSET = 490 TEST_OFFSET = df_train.last_valid_index() + 1 from MLP import MLP from Model import Model from var_model import VAR as VAR from FAVAR import FAVAR results = {} favar = FAVAR(factors_q) from MLP import MLP cols = ['CPI'] y_col = ['CPI'] X_train, y_train = df_train[cols], df_train[y_col] X_test, y_test = df_test[cols], df_test[y_col] lags = 13 def gen_x(X, start): X = VAR.gen_X(X, lags, start) # X = X[:, 1:] # remove constant