Table 7. Fit statistics and forecasting results of ARIMA models for Korean red pine and Japanese larch

Fit metric	Korean red pine		Japanese larch
	ARIMA(2, 1, 1)	ARIMA(2, 1, 2)	ARIMA(1, 1, 5)	ARIMA(6, 1, 1)
Stationary R-squared	0.65	0.65	0.39	0.43
R-squared	0.43	0.44	0.37	0.404
RMSE	99,892	105,908	73,595	78,198
MAPE (%)	14.0	14.1	8.6	8.4
MaxAPE (%)	34.146	34.539	20.7	20.5
MAE	72,359	72,799	37,106	36,313
MaxAE	139,715	137,688	102,167	101,069
Normalized BIC	23.81	24.13	23.79	24.11
Predicted volume for 2030 (m3)	745,855	744,135	463,841	362,797
UCL for 2030	994,841	1,010,000	700,112	746,673
LCL for 2030	496,868	473,804	227,569	–21,079

ARIMA: autoregressive integrated moving average, RMSE: root mean square error, MAPE: mean absolute percentage error, MaxAPE: maximum absolute percentage error, MAE: mean absolute error, MaxAE: maximum absolute error, BIC: Bayesian information criterion, UCL: upper control limit; LCL: lower control limit.