*Department of Mathematics, Nazareth Margoschis college, Nazreth, Tamilnadu, India - 628 617
**Research Scholar, Manonmaniam Sundaranar University, Tirunelveli (Department of Mathematics, B.C.M.W.G. Polytehnic, Ettayapuram-628 902) Tamil Nadu, India
***Department of Mathematics, Sri S.R.N.M. College, Sattur-626 203, Tamilnadu, India
AMS Mathematics Subject Classification (2010): 03E72, 62M10, 90C40
During the last decade different models have been designed and developed. Vamitha et.al (2012) proposed a fuzzy time series model to predict the temperature from correlated categorical data sequence obtained from similar source. In this paper, we propose a new fuzzy time series model called two factors time variant fuzzy time series forecasting based on higher order multivariate Markov chain. Two numerical data sets namely temperature and cloud density of Taipei is selected to illustrate the proposed methodand compare with some of the existing fuzzy time series methods. The proposed method gets higher average forecasting accuracy rate.
Time variant, Multivariate Markov chain, Categorical data sequence, Higher orderfuzzy time series
