Several quite severe droughts occurred in Europe in the 21st century; three of them (2003, 2012 and 2015) hit also Slovakia. The Standardized Precipitation Index (SPI) and Standardized Precipitation and Evapotranspiration Index (SPEI) were used for assessment of meteorological drought occurrence. The research was established on discharge time series representing twelve river basins in Slovakia within the period 1981–2015. Sequent Peak Algorithm method based on fixed threshold, three parametric Weibull and generalized extreme values distribution GEV, factor and multiple regression analyses were employed to evaluate occurrence and parameters of hydrological drought in 2003, 2011–2012 and 2015, and the relationship among the water balance components. Results showed that drought parameters in evaluated river basins of Slovakia differed in respective years, most of the basins suffered more by 2003 and 2012 drought than by the 2015 one. Water balance components analysis for the entire period 1931–2016 showed that because of contin
This paper presents a survey of recent successful algorithms for blind separation of determined instantaneous linear mixtures of independent sources such as natural speech or biomedical signals. These algorithms rely either on non-Gaussianity, nonstationarity, spectral diversity, or on a combination of them. Performance of the algorithms will be demonstrated on separation of a linear instantaneous mixture of audio signals (music, speech) and on artifact removal in electroencephalogram (EEG).
The purpose of this paper is twofold. Firstly, to investigate the merit of estimating probability density functions rather than level or classification estimations on a one-day-ahead forecasting the task of the silver time series.
This is done by benchmarking the Gaussian mixture neural network model (as a probability distribution predictor) against two other neural network designs representing a level estimator (the Mulit-layer perceptron network [MLP]) and a classification model (Softmax cross entropy network model [SCE]). In addition, we also benchmark the results against standard forecasting models, namely a naive model, an autoregressive moving average model (ARMA) and a logistic regression model (LOGIT).
The second purpose of this paper is to examine the possibilities of improving the trading performance of those models by applying confirmation filters and leverage.
As it turns out, the three neural network models perform equally well generating a recognisable gain while ARMA benchmark model, on the other hand, seems to have picked up the right rhythm of mean reversion in the silver time series, leading to very good results. Only when using more sophisticated trading strategies and leverage, the neural network models show an ability to identify successfully trades with a high Sharpe ratio and outperform the ARMA model.