Development of the Panel ARDL by Adding Space-Time effect to Modeling Monthly Paddy Producer Price in Java

Abstract Classic panel data modelling has large cross section (N) and small time series (T). The aim of the study was to develop a type T > N by adding space-time effect ARDL model. basic idea combine between AR (Autoregressive), DL (Distributed Lag), ST (space-time) effect. model applied paddy producer price at farmer level in Java from January 2016 December 2019 where explanatory variable Farmers’ Terms Trade. Both variables were stationary first-difference I (1). results showed that ST-ARDL good for types. with reparameterization able overcome problem multicollinearity. improve performance classic which reduce RMSE increased R 2 -adj. linear combination this cointegrated or had long-term equilibrium relationship. Another result provided better estimation than (p), (p, q) GSTAR λ ) models smaller MAPE values. For further research, can be developed interaction.