Modelling of cayenne production in Central Java using ARIMA-GARCH

Construction of ANFIS Model Based on LM-Test for Forecasting of Chili Price Data in Semarang

“…The NN architecture which applied in ANFIS consist of 5 fixed-layers [9,17,18]. Without loss of generality, the architecture of ANFIS for modeling time series data given 2 input variables −1 , −2 and single output variable t Z by assuming rule-bases of Sugeno order-one with 2 rules is as follows:…”

“…Adding input is evaluated based on LM test procedure to get a model with optimal inputs. If given 3 input variables −1 , −2 , −3 with 2 number of clusters, then the restricted model for this case can be written as [17,18]:…”

“…The number of membership function of each input can be added by using LM-test when optimal inputs of ANFIS had been selected. If given 2 input variables −1 , −2 with 2 clusters, then the restricted model can be formulated as [17,18]:…”

“…There are possible combination of rules to be generated when given membership functions and inputs. For instance, given 3 inputs −1 , −2 , −3 with 2 number of clusters, and one output by assuming first order Sugeno fuzzy system with 2 rules [17,18] 23 −3 + 20 ; then the output of ANFIS is given by equation:…”