Search

Abstract Purpose The Group Method of Data Handling (GMDH) neural network has demonstrated good performance in data mining, prediction, and optimization. Scholars have used it to forecast stock and real estate investment trust (REIT) returns in some countries and region, but not in the United States (US) REIT market. The primary goal of this study is to predict the US REIT market using GMDH and then compare its accuracy with that derived from the traditional prediction method. Design/methodology/approach To forecast the return on the US REIT index, this study used the GMDH neural network and the generalized autoregressive conditional heteroscedasticity (GARCH) model. In this test, the training samples, testing samples, and kernel functions of the GMDH model are controlled to investigate their impact on the accuracy of the machine learning approach. Corresponding experiments were performed using the GARCH model, and the accuracies of these two approaches were compared. Findings Compared with GARCH, GMDH’s accuracy is much higher, indicating that the machine learning approach can provide a highly accurate prediction of REIT prices. The size of the training samples and the kernel functions in the GMDH model affect the accuracy of the prediction results. In particular, the kernel function has a significant impact on prediction accuracy. The linear and linear covariance kernel functions are simple to train and yield accurate predictions, whereas the quadratic function is difficult to train. Even with small training samples, GMDH can outperform GARCH in prediction accuracy. Research limitations/implications Although GMDH shows good performance in predicting the US REIT return, it is still a black-box model, and the algorithm is difficult for financial analysts to develop and customize. The data used in this study come from the US REIT market, which is the world’s largest and most liquid market. Social implications This research shows that the GMDH model outperforms the GARCH model in forecasting REIT returns. Hence, investors can use the machine learning approach to make more accurate predictions of the target REITs’ returns and thus better investment decisions. Future investors and researchers may use GMDH to forecast the performance of REITs in other markets. Originality/value This is the first study to apply the GMDH neural network to the US REIT market and determine the impact of the two factors on its performance. For example, this research first discusses the impact of kernel functions on the US REIT market using the GMDH neural network. It also includes short-term daily prediction returns that were not previously considered, making it a valuable reference for financial industry analysts.

This paper presents a visualisation tool (ECvis) that aids the development of population based numerical optimisation algorithms such as genetic algorithms and differential evolution. The tool provides a simple interface with three modes: A Density mode that allows the user to quickly view the distribution and density of the population throughout the fitness and search space for high dimensional problems. This provides the ability to quickly establish where the population is clustering, which can indicate potential local and global minima; A Statistical mode that allows the user to visualise the individuals that have statistical significance in the population and their location in the search space; A Ranges mode that provides the user with a windowed average of the minimum, maximum and median value of each parameter of the population and whether the range of each parameter has changed since the previous window. This allows the user to see whether the population is exhibiting exploration or exploitation properties, as well as convergence properties such as the variance of each parameter. As examples of the usefulness of ECvis, two well known, high dimensional functions are optimised using differential evolution with ECvis being used to provide information on the performance of the optimisation.

Stock market volatility has a significant impact on many economic and financial activities in the world. Forecasting stock price movement plays an important role in setting an investment strategy or determining the right timing for trading. However, stock price movements are noisy, nonlinear, and chaotic. It is difficult to forecast stock trends for improving return on investment. Here, we proposed a novel improved particle swarm optimization (IPSO) and long-short term memory (LSTM) hybrid model for stock price forecasting. An adaptive mutation factor was used as a parameter for model optimization to avoid premature convergence to a local optimum. Furthermore, we presented a nonlinear approach to improve the inertia weight of the particle swarm optimization and then used the IPSO model to optimize the hyperparameters of LSTM. The experimental results showed that the proposed model outperformed the other related baseline models: support-vector regression, LSTM and PSO-LSTM on the Australian stock market index. These results indicated the proposed model possesses high reliability and good forecasting capability.

Halogenated MXenes have been experimentally demonstrated to be promising two-dimensional materials for a wide range of applicability. However, their physicochemical properties are largely unknown at the atomic level. In this study, we applied density functional theory (DFT) to theoretically investigate the halogenation effects on the structural, electronic, and mechanical characteristics of Ti _3 C _2 , which is the most studied MXene material. Three atomic configurations with different adsorption sites for four kinds of halogen terminals (fluorine, chlorine, bromine, and iodine) were considered. Our DFT results reveal that the adsorption site of terminals has a considerable impact on the properties of MXene. This can be ascribed to the different coordination environments of the surface Ti atoms, which change d-orbital splitting configurations of surface Ti atoms and the stabilities of systems. According to the density of states, crystal orbital Hamilton population, and charge analyses, all the considered halogenated MXenes are metallic. The electronic and mechanical properties of the halogenated MXenes are strongly dependent on the electronegativity of the halogen terminal group. The Ti–F bond has more ionic characteristics, which causes Ti _3 C _2 F _2 mechanically behave in a more ductile manner. Our DFT results, therefore, suggest that the physicochemical properties of MXenes can be tuned for practical applications by selecting specific halogen terminal groups.

Cerebral microbleeds (CMB) are increasingly present with aging and can reveal vascular pathologies associated with neurodegeneration. Deep learning-based classifiers can detect and quantify CMB from MRI, such as susceptibility imaging, but are challenging to train because of the limited availability of ground truth and many confounding imaging features, such as vessels or infarcts. In this study, we present a novel generative adversarial network (GAN) that has been trained to generate three-dimensional lesions, conditioned by volume and location. This allows one to investigate CMB characteristics and create large training datasets for deep learning-based detectors. We demonstrate the benefit of this approach by achieving state-of-the-art CMB detection of real CMB using a convolutional neural network classifier trained on synthetic CMB. Moreover, we showed that our proposed 3D lesion GAN model can be applied on unseen dataset, with different MRI parameters and diseases, to generate synthetic lesions with high diversity and without needing laboriously marked ground truth.

Autonomously spiking dopaminergic neurons of the substantia nigra pars compacta (SNpc) are exquisitely specialized and suffer toxic iron-loading in Parkinson's disease (PD). However, the molecular mechanism involved remains unclear and critical to decipher for designing new PD therapeutics. The long-lasting (L-type) CaV1.3 voltage-gated calcium channel is expressed at high levels amongst nigral neurons of the SNpc, and due to its role in calcium and iron influx, could play a role in the pathogenesis of PD. Neuronal iron uptake via this route could be unregulated under the pathological setting of PD and potentiate cellular stress due to its redox activity. This Commentary will focus on the role of the CaV1.3 channels in calcium and iron uptake in the context of pharmacological targeting. Prospectively, the audacious use of artificial intelligence to design innovative CaV1.3 channel inhibitors could lead to breakthrough pharmaceuticals that attenuate calcium and iron entry to ameliorate PD pathology.

BACKGROUND: Accumulated biological research outcomes show that biological functions do not depend on individual genes, but on complex gene networks. Microarray data are widely used to cluster genes according to their expression levels across experimental conditions. However, functionally related genes generally do not show coherent expression across all conditions since any given cellular process is active only under a subset of conditions. Biclustering finds gene clusters that have similar expression levels across a subset of conditions. This paper proposes a seed-based algorithm that identifies coherent genes in an exhaustive, but efficient manner. METHODS: In order to find the biclusters in a gene expression dataset, we exhaustively select combinations of genes and conditions as seeds to create candidate bicluster tables. The tables have two columns (a) a gene set, and (b) the conditions on which the gene set have dissimilar expression levels to the seed. First, the genes with less than the maximum number of dissimilar conditions are identified and a table of these genes is created. Second, the rows that have the same dissimilar conditions are grouped together. Third, the table is sorted in ascending order based on the number of dissimilar conditions. Finally, beginning with the first row of the table, a test is run repeatedly to determine whether the cardinality of the gene set in the row is greater than the minimum threshold number of genes in a bicluster. If so, a bicluster is outputted and the corresponding row is removed from the table. Repeating this process, all biclusters in the table are systematically identified until the table becomes empty. CONCLUSIONS: This paper presents a novel biclustering algorithm for the identification of additive biclusters. Since it involves exhaustively testing combinations of genes and conditions, the additive biclusters can be found more readily.