Training Multilayer Perceptron with Genetic Algorithms and Particle Swarm Optimization for Modeling Stock Price Index Prediction

Predicting stock market (SM) trends is an issue of great interest among researchers, investors and traders since the successful prediction of SMs’ direction may promise various benefits. Because of the fairly nonlinear nature of the historical data, accurate estimation of the SM direction is a rather challenging issue. The aim of this study is to present a novel machine learning (ML) model to forecast the movement of the Borsa Istanbul (BIST) 100 index. Modeling was performed by multilayer perceptron–genetic algorithms (MLP–GA) and multilayer perceptron–particle swarm optimization (MLP–PSO) in two scenarios considering Tanh (x) and the default Gaussian function as the output function. The historical financial time series data utilized in this research is from 1996 to 2020, consisting of nine technical indicators. Results are assessed using Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE) and correlation coefficient values to compare the accuracy and performance of the developed models. Based on the results, the involvement of the Tanh (x) as the output function, improved the accuracy of models compared with the default Gaussian function, significantly. MLP–PSO with population size 125, followed by MLP–GA with population size 50, provided higher accuracy for testing, reporting RMSE of 0.732583 and 0.733063, MAPE of 28.16%, 29.09% and correlation coefficient of 0.694 and 0.695, respectively. According to the results, using the hybrid ML method could successfully improve the prediction accuracy.     

Approximate analysis and design of rectangular-lattice photonic crystals

We present a theoretical analysis of wave propagation in rectangular- and square-lattice photonic crystals by approximating the arbitrary refractive-index function with a staircase profile. This profile has a number of tunable parameters that allow one to fit the band structures of the staircase and the desired structure over a large frequency range. The staircase profile is such that its corresponding two-dimensional wave equation decomposes into two one-dimensional equations by means of a constant decoupling parameter. We show that basic features of the band structure and Bloch waves in photonic crystals can be analyzed theoretically through this simple approximation.     

On the calibration of the Chaboche hardening model and a modified hardening rule for uniaxial ratcheting prediction

A systematic mathematical approach is developed in the context of uniaxial cyclic ratcheting for the parameter determination of the decomposed Chaboche hardening rule. This is achieved by deriving the relation between the evolution of the backstress and the plastic strain accumulation. Unlike current calibration techniques where a trial–error approach is employed to fit the simulation results to experimental data, the proposed method determines the parameters directly from uniaxial ratcheting experiments. Numerical results indicate that Chaboche’s hardening model is much more efficient than what has been demonstrated before. Finally, as an improvement to the decomposed model, a modification is made to one of the backstress components. This improved component enables the model to predict uniaxial ratcheting with more accuracy.     

Microbundles, manifolds and metrisability

The notion of a microbundle was introduced in the 1960s but the theory came to an abrupt halt when it was shown that for a metrisable manifold, microbundles are equivalent to fibre bundles. In this paper we consider microbundles over non-metrisable manifolds. In some cases microbundles are equivalent to fibre bundles but in others they are not. In particular, we show that a manifold is metrisable if and only if its tangent microbundle is equivalent to a fibre bundle. We also illustrate that for some non-metrisable manifolds every trivial microbundle contains a trivial fibre bundle whereas other manifolds may support a trivial microbundle not containing a trivial fibre bundle.

     

An automatic phase error compensating multiphase LC oscillator: analysis and design

In this work a self phase error compensating multiphase LC oscillator is proposed. Mismatches in LC tanks are considered as the main source of phase errors. Considering this, an analytical approach is proposed to find a relationship between phase errors and their corresponding coupling factors as a necessary condition for phase error cancellation then a self compensating circuit is proposed accordingly. The compensation circuit first detects the phase errors then employs them in a current tuner to change each stage coupling factor to reduce phase errors. The building blocks of the proposed circuit are investigated and the transistor level implementation of each one is presented afterwards. The theoretical results are evaluated and confirmed through simulations using ADS software in 0.18 μm CMOS technology. Simulation results show that not only phase errors are reduced with respect to the previous methods but also this procedure does not impose any further power consumption and phase noise to the circuit.     

One-pot,three-component,catalyst-free synthesis of novel derivatives of pyrido-[2,3-d]pyrimidines under ultrasonic irradiations

Novel derivatives of pyrido-[2,3-d] pyrimidines were synthesized via an efficient and catalyst- free, one-pot, three-component reaction of 3-(1-methyl-1H-pyrrol-2-yl)-3-oxopropanenitrile, 6-amino-(ethylthio)pyrimidine-4(3H)-one, and arylaldehydes under ultrasonic irradiations. This practical method furnished the desired pyridopyrimidines in high yields (79–89%) and short reaction times (25–40 min).