Fuzzy Autocorrelation Model with Confidence Intervals of Fuzzy Random Data

Abstract

Economic analyses are typical methods based on time-series data or cross-section data. Economic systems are complex because they involve human behaviors and are affected by many factors. When a system includes such uncertainty, as those concerning human behaviors, a fuzzy system approach plays a pivotal role in such analysis.

In this paper, we propose a fuzzy autocorrelation model with confidence intervals of fuzzy random time-series data. These confidence intervals play an essential role in dealing with fuzzy random data on the fuzzy autocorrelation model that we have presented. We analyze tick-by-tick data of stock transactions and compare two time-series models, a fuzzy autocorrelation model proposed by us, and a new fuzzy time-series model that we propose in this paper.