Table 8 Comparison of proposed method with statistical method
From: Mining precise cause and effect rules in large time series data of socio-economic indicators
Dataset | Indicators relationships | Extracted rules | Statistical methods | ||
|---|---|---|---|---|---|
Proposed method | Granger causality | Bayesian network | |||
Synthetic-1 (I1–I6) | Binary | I1 → I3 | ✓ | ✓ | ✓ |
Many to one | (I2, I4) → I5 | ✓ | |||
Transitive | I1 → I3 → I6 | ✓ | ✓ | ||
Cyclic | I1 ←→ I3 | ✓ | |||
Synthetic-2 (I1–I10) | Binary | I1 → I7, I2 → I7, I7 → I2, I1 → I3, I7 → I8 | ✓ | ✓ | ✓ |
Many to one | (I6, I9) → I7 | ✓ | |||
Transitive | I1 → I7 → I8 | ✓ | ✓ | ||
Cyclic | I2 ←→ I7 | ✓ | |||
WTO | Binary | Chemicals → Textiles Chemicals → OTE | ✓ | ✓ | ✓ |
Many to one | (OTE, Textiles) → EDOE | ✓ | |||
Transitive | IS → OM → ICEC | ✓ | ✓ | ||
Cyclic | OM ←→ IS | ✓ | |||
IMF | Binary | GGR → VEG | ✓ | ✓ | ✓ |
Many to one | (GGR, GNS) → TI | ✓ | |||
Transitive | GDP → VIG → TI | ✓ | ✓ | ||
Cyclic | CAB ←→ VEGS | ✓ | |||
World Bank data | Binary | CP → ARME | ✓ | ✓ | ✓ |
Many to one | (FDI, FR) → CPI | ✓ | |||
Transitive | AR → AG → CO2 | ✓ | ✓ | ||
Cyclic | GDP ←→ CY | ✓ | |||