###################################################################
###### find missing dates #########################################
###################################################################
vector_dates = np.concatenate([
np.array(data[data.columns[0]]),
np.array(data2[data2.columns[0]]),
np.array(data3[data3.columns[0]]),
np.array(data4[data4.columns[0]]),
np.array(data5[data5.columns[0]]),
np.array(data6[data6.columns[0]])
])
global_dates = temp.dates(pd.Series(vector_dates))
missing = Functions_for_TSP.finding_missing_dates(global_dates, milano)
missing2 = Functions_for_TSP.finding_missing_dates(global_dates, torino)
test = Functions_for_TSP.update_meteo(milano, torino)
test.to_csv('storico_milano_aggiornato.txt', sep="\t", index=False)
missing2 = Functions_for_TSP.finding_missing_dates(global_dates, test)
missingca = Functions_for_TSP.finding_missing_dates(global_dates, ca)
missingpa = Functions_for_TSP.finding_missing_dates(global_dates, pa)
missingrc = Functions_for_TSP.finding_missing_dates(global_dates, rc)
missingfi = Functions_for_TSP.finding_missing_dates(global_dates, fi)
diffmiro = Functions_for_TSP.simulate_meteo(milano, roma)
difffiro = Functions_for_TSP.simulate_meteo(fi, roma)
diffparo = Functions_for_TSP.simulate_meteo(pa, roma)
diffcaro = Functions_for_TSP.simulate_meteo(ca, roma)
plt.plot(min_season) D, Y = temp.create_dataset(data, "ven") Y = np.array(Y) acf, Q, P, = statsmodels.tsa.stattools.acf(Y, nlags=48, qstat=True) statsmodels.graphics.tsaplots.plot_acf(Y, lags=1000) per = statsmodels.tsa.stattools.periodogram(Y) plt.plot(per) S_per = pd.Series(per) S_per.describe() peaks = Functions_for_TSP.find_peaks(per, 10) FE = Functions_for_TSP.fourierExtrapolation(Y, n_predict=24) fitted_FE = FE[0:8736] diff = Y - fitted_FE np.mean(diff) np.var(diff) #RMSE = np.sqrt(np.mean(diff**2)) sp_y = Functions_for_TSP.Signum_Process(Y) sp_f = Functions_for_TSP.Signum_Process(fitted_FE) sp_p = sp_y * sp_f
