Автокорреляция в глобальном стохастическом тренде
Korhonen and Peresetsky (2013) suggested a new Kalman-filter type model of financial markets to extract a global stochastic trend from discrete non-synchronous data on daily stock market index returns from different markets. We extend this model to allow the correlation between increments of this global trend on neighbor intervals. Existence of that non-zero correlation is demonstrated. However it does not mean that it helps forecast daily returns of the stock indices itself, since the global stochastic trend is unobservable. Forecasting performance of the model with three stock markets is explored.