Integration of genetic fuzzy systems and artificial neural networks for stock price forecasting

Stock market prediction is regarded as a challenging task in financial time-series forecasting. The central idea to successful stock market prediction is achieving best results using minimum required input data and the least complex stock market model. To achieve these purposes this article presents an integrated approach based on genetic fuzzy systems (GFS) and artificial neural networks (ANN) for constructing a stock price forecasting expert system. At first, we use stepwise regression analysis (SRA) to determine factors which have most influence on stock prices. At the next stage we divide our raw data into k clusters by means of self-organizing map (SOM) neural networks. Finally, all clusters will be fed into independent GFS models with the ability of rule base extraction and data base tuning. We evaluate capability of the proposed approach by applying it on stock price data gathered from IT and Airlines sectors, and compare the outcomes with previous stock price forecasting methods using mean absolute percentage error (MAPE). Results show that the proposed approach outperforms all previous methods, so it can be considered as a suitable tool for stock price forecasting problems.     

Modeling metallurgical responses of coal Tri-Flo separators by a novel BNN:a "Conscious-Lab" development

Tri-flo cyclone,as a dense-medium separation device,is one of the most typical environmentally friendly industrial techniques in the coal washery plants.Surpris...     

Study of Adsorption/Desorption Effect of 2-Mercaptobenzothiazole as Sweet Corrosion Inhibitor on API-5L X60 Steel

Protection of Metals and Physical Chemistry of Surfaces - In this study, the inhibitive performance of 2-mercaptobenzothiazole (2MBT) and 2-aminobenzothiazole (2ABT) were investigated on API-5L X60...     

Effect of chemical environments on palladium phthalocyanine thin film sensors for humidity analysis

Chemiresistors based on palladium phthalocyanine (PdPc) thin films were investigated as humidity sensors. The samples were thermally evaporated onto gold electrodes with a thickness about 100 nm. Optical and electrical characteristics of PdPc thin films were investigated by scanning electron microscopy (SEM), X-ray diffraction (XRD), and electrical measurements. The SEM image demonstrates PdPc (30–60 nm) nanosized particles, and XRD pattern shows that thin films are in α-phase at room temperature. Electrical measurements also confirm that PdPc exhibit semiconducting and photoconducting behaviors, and thermal activation energies of thin films were calculated. After that, the sensitivity and reversibility of devices were investigated on exposure to 20–90% RH in various chemical environments at 293 and 323 K. The response time (35–45 s) and recovery time (75–105 s) of sensors were measured at 293 K with respect to different chemical environments. At last, the stability of devices versus different RH% and chemical environments were tested. The sensors show very good stability on exposure to RH for a period of 2 months but their stability has been reduced in ethanol, acetone, and ammonia environments.     

Interfacial Reactions Between BaAl2Si2O8 and Molten Al Alloy at 850°C

A corrosion cup test was undertaken using BaAl₂Si₂O₈ and Al4.1Zn3.2Mg alloy, heated in air for 150 h at 850°C. Electron probe microanalysis, X‐ray diffraction, and scanning electron microscopy coupled with energy dispersive spectroscopy were used to identify the mineralogical and microstructural changes at the interfaces. The microstructural results revealed three microstructural areas: (1) Spinel layer with large numbers of Al alloy channels; (2) interfacial area with mainly alumina, spinel, and BaAl₂Si₂O₈; and (3) interdiffusion zone chemically close to barium hexaaluminate. The principal observations are:

相似文献   

Improving Blood Compatibility of Polyethersulfone Hollow Fiber Membranes via Blending with Sulfonated Polyether Ether Ketone

Appropriate membrane for blood contacting applications requires hemocompatibility and high permeation flux; it should inhibit proteins or platelets adsorption and still possess high permeability. Aiming to improve the polyethersulfone (PES) hollow fiber membrane hemocompatibility, sulfonated polyether ether ketone (SPEEK) is self‐synthesized in the present research and added to PES in different ratios. Scanning electron microscopy images have revealed significant changes in PES membranes structure after addition of SPEEK, which can influence water permeation property of the membranes. Water contact angles of the membranes have reduced from 75° to 50° after addition of 4 wt% SPEEK. Influence of SPEEK addition on hemocompatibility of the PES membranes is evaluated via protein (bovine serum albumin) adsorption, platelet attachment, and coagulation time (APTT and TT) assays. Obtained results reveal that hemocompatibility of the modified hollow fiber membranes is enhanced as a result of emerging repulsive forces between negative charges on the membranes surface and negatively charge blood components.

     

Screening of Food Samples for Zearalenone Toxin Using an Electrochemical Bioassay Based on DNA–Zearalenone Interaction

This study was conducted to design a biosensor as a new, rapid, and sensitive tool for investigation of binding of zearalenone with double-stranded DNA (dsDNA). Polydiallyldimethylammonium chloride (PDDA) as a polycation and multiwall carbon nanotubes (MWCNTs) provide a positively charged surface with a high surface area for the immobilization of dsDNA as a polyanion on the surface of pencil graphite electrode (PGE). Using the dsDNA/MWCNT–PDDA-modified PGE, it was possible to detect the interaction of zearalenone with dsDNA, which allowed us to apply the dsDNA-modified electrode for trace determination of zearalenone. The changes at the oxidation signal of adenine were evaluated before/after each modification/immobilization step. By using dsDNA/PDDA–MWCNT/PGE, zearalenone could be detected as low as 0.005 ng mL^?1. The relative standard deviation of five measurements of 0.5 ng mL^?1 zearalenone was found to be 4.2 %. Finally, the highly stable electrochemical biosensor was applied to analyze the zearalenone concentration in milk and wheat samples.

Open image in new window

Graphical Abstract xElectrochemical DNA Biosensor for Zearalenone detection

     

Adoption and use of e-commerce in SMEs

In this study, nonlinear Logit and probit models are used to analyse the important factors that impact on the tendency of small- and medium-sized enterprises (SMEs) to use electronic commerce in one of the industrial parks in Kermanshah province of Iran. Furthermore, it identifies the major barriers of e-commerce adoption in these SMEs. The Sample of the study includes 35 SMEs in this province. The results of probit and Logit models suggest that a lower level of the internet service costs, motivates firms to adopt e-commerce. In addition, if the government provides free e-commerce facilities for SMEs, it can encourage SMEs to adopt e-commerce. Furthermore, the results show that due to uncertainty in the e-commerce, producing high-quality products and traditional exports cannot raise the tendency of sample firms to adopt e-commerce.     

On the design of feedback control for relative stability

A basic theorem on relative stability is generalized and carefully interpreted. Two new theorems are then presented which simplify the calculation of feedback gains to achieve a prescribed damping ratio.