Table 5 Results of multi-step-ahead... | Emerald Publishing

Table 5

Results of multi-step-ahead forecasting for Jiuzhaigou

Period	Pre-COVID-19			Post-COVID-19
Measure	MAE	MAPE	RMSE	MAE	MAPE	RMSE
MGASF–CNN–LSTM	552.543	0.053	2119.660	809.310	0.341	2756.526
MGADF–CNN–LSTM	576.858	0.065	2516.128	799.318	0.257	2660.791
MMTF–CNN–LSTM	554.754	0.054	2229.129	811.480	0.461	2761.109
MRP–CNN–LSTM	567.366	0.054	2461.164	789.101	0.214	2555.037
GASF–CNN–LSTM	4977.555	1.981	7375.662	5450.734	2.718	7286.421
GADF–CNN–LSTM	4805.180	1.917	7286.421	5611.4971	2.909	7501.466
MTF–CNN–LSTM	4941.158	1.956	7320.211	5647.1357	3.084	7535.970
RP–CNN–LSTM	4965.838	1.958	7365.759	5529.1962	2.893	7375.662
ARIMA	6549.528	14.430	8325.730	6688.989	14.508	8480.441
Holt–Winters	6537.387	14.313	8305.170	6902.982	15.711	8599.571
OLS	2148.602	0.802	3459.231	2786.447	1.088	3776.166
KNN	1465.074	0.493	2832.017	1952.037	0.541	2954.167
XG-BOOST	1502.098	0.507	2861.728	2028.506	0.556	3268.129

Period	Pre-COVID-19			Post-COVID-19
Measure	MAE	MAPE	RMSE	MAE	MAPE	RMSE
MGASF–CNN–LSTM	552.543	0.053	2119.660	809.310	0.341	2756.526
MGADF–CNN–LSTM	576.858	0.065	2516.128	799.318	0.257	2660.791
MMTF–CNN–LSTM	554.754	0.054	2229.129	811.480	0.461	2761.109
MRP–CNN–LSTM	567.366	0.054	2461.164	789.101	0.214	2555.037
GASF–CNN–LSTM	4977.555	1.981	7375.662	5450.734	2.718	7286.421
GADF–CNN–LSTM	4805.180	1.917	7286.421	5611.4971	2.909	7501.466
MTF–CNN–LSTM	4941.158	1.956	7320.211	5647.1357	3.084	7535.970
RP–CNN–LSTM	4965.838	1.958	7365.759	5529.1962	2.893	7375.662
ARIMA	6549.528	14.430	8325.730	6688.989	14.508	8480.441
Holt–Winters	6537.387	14.313	8305.170	6902.982	15.711	8599.571
OLS	2148.602	0.802	3459.231	2786.447	1.088	3776.166
KNN	1465.074	0.493	2832.017	1952.037	0.541	2954.167
XG-BOOST	1502.098	0.507	2861.728	2028.506	0.556	3268.129

Source(s): Authors’ own work