Table 2 Comparison of accuracy measurements between Model 1 (local variables) and Model 2 (adds remote climate modes to Model 1) in Malaysia, Peninsular Malaysia and Sabah/Sarawak.

From: Climate Based Predictability of Oil Palm Tree Yield in Malaysia

Region

Model

ME

RMSE

MAE

MPE

MAPE

MASE

ACF1

Malaysia

1

−1.79e–13

0.699

0.544

−0.149

2.97

0.529

−0.234

2

4.10e–14

0.421

0.331

−0.050

1.76

0.322

−0.404

Peninsular Malaysia

1

7.57e–14

0.662

0.553

−0.1280

3.0

0.476

−0.0941

2

6.67e–14

0.545

0.426

−0.0858

2.3

0.367

−0.1100

Sabah/Sarawak

1

3.02e–13

1.010

0.805

−0.342

4.55

0.780

−0.0115

2

3.45e–13

0.916

0.701

−0.281

3.94

0.679

0.2250

  1. ME is the Mean Error (Mg ha−1), RMSE is the Root Mean Square Error (Mg ha−1), MAE is the Mean Absolute Error (Mg ha−1), MPE is the Mean Percentage Error (%), MAPE is the Mean Absolute Percentage Error (%), MASE is the Mean Absolute Scaled Error and ACF1 is the autocorrelation of errors at lag 1. Values are estimated from bootstrap (10,000 permutations).
Back to article page