Linguistic Descriptions As a Modeling Tool For Multivariate Time Series

We propose linguistic associations mining as a technique to create the models of the multivariate time series. We define various linguistic evaluative expressions on the range of the values of the time series and variables derived from them. We mine linguistic associations then and interpret them as IF-THEN rules in the framework of Perception based Logic Deduction (PbLD). The mined rules provide the linguistic descriptions of various relationships between the time series. We showcase our suggested methodology in a macroeconomic example where we compare our approach with Dynamic Stochastic General Equilibrium (DSGE) model, that is frequently used in the macroeconomic modeling.