SCOT Modeling and Its Statistical Applications of Time Series

We modeled and applied Stochastic COntext Tree (SCOT) for statistical inference about financial, literary and seismological stationary strings. We analyzed several models viewed as simplified approaches to financial modeling: evaluate their stationary distribution, entropy rate and convergence to the Brownian motion. We also estimated the SCOT parameters and tested homogeneity of data strings using additive state functions of SCOT trajectories. Moreover, we justified properties of the homogeneity test statistic introduced there and for finding active inputs of sparse systems with correlated noise.