The paper describe the North-East volatility wind effect in financial time series. This effect explains the volatility of transition from low frequency to high frequency signal components. The signal in this case is understood pre-processed financial time series. North-East wind volatility effect has been studied by filtering the signal using direct and inverse transforms continuous wave (Direct and Inverse Continuous Wavelet Transform). Object of investigation is logarithmic variance (volatility indicator) and its evolution in time, as well as the volatility transimission between signal components (between volatility layers). As a result, sophisticated nature of volatility and its transmission between volatility layers is discovered. Mechanism is ilustrated in a complex plane. This brings out better understanding of North-East of volatility wind effect and volatility evolution nature. North-East volatility wind effect in the complex plane provides so-called implosion South-West wind volatility effect provides explosion. North-East of volatility wind effect is discussed, research algorithms and results are provided. Work explores wave nature of volatility in financial markets, a new opportunities in volatility long-term prediction are shown. Predictions involve wavelet decomposition, neural networks, complex transformations. PhD thesis describes opportunity to reveal and trace financial market instability on early stages (at least 2 years before crisis). The developed model allows to predict the non-stationary time series.