Kristina Lukoseviciute

Evolutionary algorithms for the selection of time lags for time series forecasting by fuzzy inference systems

Time series forecasting by fuzzy inference systems based on optimal non-uniform attractor embedding in the multidimensional delay phase space is analyzed in this paper. A near-optimal set of time lags is identified by evolutionary algorithms after the optimal dimension of the reconstructed phase space is determined by the FNN (false nearest neighbors) algorithm. The fitness function is constructed in such a way that it represents the spreading of the attractor in the delay coordinate space but does not contain any information on prediction error metrics. The weighted one-point crossover rule enables an effective identification of near-optimal sets of non-uniform time lags which are better than the globally optimal set of uniform time lags. Thus the reconstructed information on the properties of the underlying dynamical system is directly elaborated in the fuzzy prediction system. A number of numerical experiments are used to test the functionality of this method.

Non-uniform attractor embedding for time series forecasting by fuzzy inference systems

A new method for identification of an optimal set of time lags based on non-uniform attractor embedding from the observed non-linear time series is proposed in this paper. Simple deterministic method for the determination of non-uniform time lags comprises the pre-processing stage of the time series forecasting algorithm which is implemented in the form of a fuzzy inference system. Identification of embedding parameters of the underlying dynamical system includes not only optimization of time lags but also determination of optimal dimension of the reconstructed phase space. Experiments done with benchmark chaotic time series show that the proposed method can considerably improve the forecasting accuracy. The proposed method seems to be an efficient candidate for prediction of time series with multiple time scales and noise.

Short-term time series forecasting based on the identification of skeleton algebraic sequences

A new short-term time series forecasting method based on the identification of skeleton algebraic sequences is proposed in this paper. The concept of the rank of the Hankel matrix is exploited to detect a base fragment of the time series. Particle swarm optimization and evolutionary algorithms are then used to remove the noise and identify the skeleton algebraic sequence. Numerical experiments with an artificially generated and a real-world time series are used to illustrate the functionality of the proposed method.

Short-term time series algebraic forecasting with mixed smoothing

Short-term time series algebraic prediction technique with mixed smoothing is presented in this paper. Evolutionary algorithms are employed for the identification of a near-optimal algebraic skeleton from the available data. Direct algebraic predictions are conciliated by internal errors of interpolation and external differences from the moving average. Computational experiments with real world time series are used to demonstrate the effectiveness of the proposed forecasting algorithm.

Algebraic segmentation of short nonstationary time series based on evolutionary prediction algorithms

Algebraic segmentation of short nonstationary time series is presented in this paper. The proposed algorithm is based on the algebraic one step-forward predictor which is used to identify a temporal near-optimal algebraic model of the real-world time series. A combinatorial algorithm is used to identify intervals where prediction errors are lower than a predefined level of acceptable accuracy. Special deterministic strategy is developed for the selection of this acceptable level of prediction accuracy and is individually determined for every time series. The nonparametric identification of quasistationary segments is performed without the employment of any statistical estimator. Several standard real-world time series are used to demonstrate the efficiency of the proposed technique.

Investigation of the Time of Flight estimation errors induced by neighboring signals

Investigation of the errors of Time of Flight (ToF) estimate of ultrasonic signals in case of multiple reflections environment is presented. Multiple signals environment induces neighboring signal energy leak into analyzed signal. This in turn introduces some errors in ToF estimate. Narrow signals like single pulse, should posses lower influence that CW burst or chirp signals. Then high energy signals (burst or chirp) will always be jeopardized in case of multiple reflections. New class of SS signals, trains of pulses of arbitrary pulse width and position (APWP), is expected to have properties similar to chirp (wide, controllable bandwidth) and the single pulse (low correlation sidelobes) signals. Goal of the investigation was to compare the multi-reflections performance in Time of Flight estimation for classical signals (rectangular pulse, step and CW burst) and spread spectrum signals (chirp and APWP). Signals from fixed delay line were recorded, aligned in time and averaged to produce noise-free reference signal which was used to simulate the plate thickness measurement and liquid velocity estimation with variable spacing of signals in time. Simulation and experimental investigation results are presented and possible problem solutions suggested.

Bernstein polynomials for adaptive evolutionary prediction of short-term time series

Abstract We introduce a short-term time series prediction model by means of evolutionary algorithms and Bernstein polynomials. This adapts Bernstein-type algebraic skeletons to extrapolate and predict short time series. A mixed smoothing strategy is used to achieve the necessary balance between the roughness of the algebraic prediction and the smoothness of the moving average. Computational experiments with standardized real world time series illustrate the accuracy of this approach to short-term prediction.

Reiterative deconvolution: New technique for time of flight estimation errors reduction in case of close proximity of two reflections

HighlightsTime of flight estimation of two interfering ultrasonic reflections is analyzed.Time of flight estimation is biased when two reflections are in close proximity.The closer the reflections the larger is the bias error.Reiterative deconvolution is proposed to reduce these errors.Experiment results indicate that significant bias error reduction is possible. Abstract ToF estimation errors for two reflections have been analyzed. Case of high signal‐to‐noise ratio was assumed, when accuracy of the order of few nanoseconds can be achieved. It was indicated that additional bias errors are introduced in ToF estimator when other reflection is in close temporal proximity to the reflection of interest. Research demonstrates that iterative deconvolution does not improve the accuracy significantly. Novel technique is suggested which addresses this issue by means of a reiterative deconvolution. Simulation and experimental performance evaluation is presented. Bias error quickly diminishes with every reiteration when using new technique and is below the other errors after few reiterations.

Bernstein polynomials for evolutionary algebraic prediction of short time series

Short time series prediction technique based on Bernstein polynomials is presented in this paper. Firstly, the straightforward Bernstein polynomial extrapolation scheme is improved by extending the interval of approximation. Secondly, the forecasting scheme is designed in the evolutionary computational setup which is based on the conciliation between the coarseness of the algebraic prediction and the smoothness of the time average prediction. Computational experiments with the test time series suggest that this time series prediction technique could be applicable for various forecasting applications.

Algebraic Level-Set Approach for the Segmentation of Financial Time Series

Adaptive algebraic level-set segmentation algorithm of financial time series is presented in this paper. The proposed algorithm is based on the algebraic one step-forward predictor with internal smoothing, which is used to identify a near optimal algebraic model. Particle swarm optimization algorithm is exploited for the detection of a base algebraic fragment of the time series. A combinatorial algorithm is used to detect intervals where predictions are lower than a predefined level. Moreover, the combinatorial algorithm does assess the simplicity of the identified near optimal algebraic model. Automatic adaptive identification of quasi-stationary segments can be employed for complex financial time series.