Comparison of fuzzy time forecasting multi-factor short cross associations with more partitions on the main factor and partitions on the second factor.

Economic decisions that have many determinants on the estimation of macroeconomic variables require accurate estimates. These estimates are very influential on the results of future decisions. However, the future is often filled with uncertainty. This uncertainty that we have can be reduced by forecasting. Forecasting have four basic steps, one of those steps is each fuzzy logic relation that has one consequent and more than one premise that reflects the relationship between fuzzy values. The high order fuzzy time series forecast method is more suitable than the first order fuzzy time series forecast since economic decisions have multifactor problems. Forecasting comes from the influence of these two factors. This prediction will be made as the main factor. In previous studies, there were many uses of multi-factor short cross-association with partitioned fuzzy logical relationships based forecasting models of time series with more partitions of the second factor. This research tries to partition more on the main factors. Researchers find out that the AFER and MSE of the modification method is lower than previous method mentioned and the line is closer to the actual line. It turns out our suggested modified forecasting method for fuzzy time series produces superior result. [ABSTRACT FROM AUTHOR]