Efficiency Comparison of Prediction Methods and Analysis of Factors Affecting Savings of People in the Central Region of Thailand
Table 5
Results of efficiency comparison in predicting savings of people in the central region by multiple linear regression analysis and optimized artificial neural network.
Household savings
Root mean square error (RMSE)
Mean absolute error (MAE)
MLR
Unoptimized ANN
MLR
Unoptimized ANN
Low-income
91,397.9315
127,910.8087
243,314.6593
54,384.4958
Middle-income
259,150.5120
326,001.4820
58,097.1327
170,355.9166
High-income
851,552.7515
2,779,219.6003
1,926,924.5832
877,077.2266
Household savings
Coefficient of determination (R2)
Processing time (second)
MLR
Unoptimized ANN
MLR
Unoptimized ANN
Low-income
0.7152
0.6807
3
8
Middle-income
0.7608
0.6934
5
52
High-income
0.7895
0.6277
4
13
Bold values indicate the better method from the comparison of the two methods.