Prediction of Parkinson’s Disease tremor onset using a radial basis function neural network based on particle swarm optimization

Deep Brain Stimulation (DBS) has been successfully used throughout the world for the treatment of Parkinson's disease symptoms. To control abnormal spontaneous electrical activity in target brain areas DBS utilizes a continuous stimulation signal. This continuous power draw means that its implanted battery power source needs to be replaced every 18–24 months. To prolong the life span of the battery, a technique to accurately recognize and predict the onset of the Parkinson's disease tremors in human subjects and thus implement an on-demand stimulator is discussed here. The approach is to use a radial basis function neural network (RBFNN) based on particle swarm optimization (PSO) and principal component analysis (PCA) with Local Field Potential (LFP) data recorded via the stimulation electrodes to predict activity related to tremor onset. To test this approach, LFPs from the subthalamic nucleus (STN) obtained through deep brain electrodes implanted in a Parkinson patient are used to train the network. To validate the network's performance, electromyographic (EMG) signals from the patient's forearm are recorded in parallel with the LFPs to accurately determine occurrences of tremor, and these are compared to the performance of the network. It has been found that detection accuracies of up to 89% are possible. Performance comparisons have also been made between a conventional RBFNN and an RBFNN based on PSO which show a marginal decrease in performance but with notable reduction in computational overhead.

Use of the statistical properties of the wavelet-transform coefficients for optimization of integration time in Fourier transform spectrometry

We show that an analysis of the mean and variance of discrete wavelet coefficients of coaveraged time-domain interferograms can be used as a specification for determining when to stop coaveraging. We also show that, if a prediction model built in the wavelet domain is used to determine the composition of unknown samples, a stopping criterion for the coaveraging process can be developed with respect to the uncertainty tolerated in the prediction.

Dynamic integrated system optimization and parameter estimation for discrete time optimal control of nonlinear systems

An algorithm for solving nonlinear discrete time optimal control problems with model-reality differences is presented. The technique uses Dynamic Integrated System Optimization and Parameter Estimation (DISOPE), which achieves the correct optimal solution in spite of deficiencies in the mathematical model employed in the optimization procedure. A version of the algorithm with a linear-quadratic model-based problem, implemented in the C+ + programming language, is developed and applied to illustrative simulation examples. An analysis of the optimality and convergence properties of the algorithm is also presented.

On combination of SMOTE and particle swarm optimization based radial basis function classifier for imbalanced problems

The combination of the synthetic minority oversampling technique (SMOTE) and the radial basis function (RBF) classifier is proposed to deal with classification for imbalanced two-class data. In order to enhance the significance of the small and specific region belonging to the positive class in the decision region, the SMOTE is applied to generate synthetic instances for the positive class to balance the training data set. Based on the over-sampled training data, the RBF classifier is constructed by applying the orthogonal forward selection procedure, in which the classifier structure and the parameters of RBF kernels are determined using a particle swarm optimization algorithm based on the criterion of minimizing the leave-one-out misclassification rate. The experimental results on both simulated and real imbalanced data sets are presented to demonstrate the effectiveness of our proposed algorithm.

The impact of exchanges rates on international real estate portfolio allocation

This paper, examines whether the asset holdings and weights of an international real estate portfolio using exchange rate adjusted returns are essentially the same or radically different from those based on unadjusted returns. The results indicate that the portfolio compositions produced by exchange rate adjusted returns are markedly different from those based on unadjusted returns. However following the introduction of the single currency the differences in portfolio composition are much less pronounced. The findings have a practical consequence for the investor because they suggest that following the introduction of the single currency international investors can concentrate on the real estate fundamentals when making their portfolio choices, rather than worry about the implications of exchange rate risk.

Real estate portfolio size and risk reduction

The reduction of portfolio risk is important to all investors but is particularly important to real estate investors as most property portfolios are generally small. As a consequence, portfolios are vulnerable to a significant risk of under-performing the market, or a target rate of return and so investors may be exposing themselves to greater risk than necessary. Given the potentially higher risk of underperformance from owning only a few properties, we follow the approach of Vassal (2001) and examine the benefits of holding more properties in a real estate portfolio. Using Monte Carlo simulation and the returns from 1,728 properties in the IPD database, held over the 10-year period from 1995 to 2004, the results show that increases in portfolio size offers the possibility of a more stable and less volatile return pattern over time, i.e. down-side risk is diminished with increasing portfolio size. Nonetheless, increasing portfolio size has the disadvantage of restricting the probability of out-performing the benchmark index by a significant amount. In other words, although increasing portfolio size reduces the down-side risk in a portfolio, it also decreases its up-side potential. Be that as it may, the results provide further evidence that portfolios with large numbers of properties are always preferable to portfolios of a smaller size.

How often does direct real estate increase the risk-adjusted performance of the US mixed-asset portfolio?

The “case for real estate” in the mixed-asset portfolio is a topic of continuing interest to practitioners and academics. The argument is typically made by comparing efficient frontiers of portfolio with real estate to those that exclude real estate. However, most investors will have held inefficient portfolios. Thus, when analysing the real estate’s place in the mixed-asset portfolio it seems illogical to do so by comparing the difference in risk-adjusted performance between efficient portfolios, which few if any investor would have held. The approach adopted here, therefore, is to compare the risk-adjusted performance of a number of mixed-asset portfolios without real estate (which may or not be efficient) with a very large number of mixed-asset portfolios that include real estate (which again may or may not be efficient), to see the proportion of the time when there is an increase in risk-adjusted performance, significant or otherwise using appraisal-based and de-smoothed annual data from 1952-2003. So to the question how often does the addition of private real estate lead to increases the risk-adjusted performance compared with mixed-asset portfolios without real estate the answer is almost all the time. However, significant increases are harder to find. Additionally, a significant increase in risk-adjusted performance can come from either reductions in portfolio risk or increases in return depending on the investors’ initial portfolio structure. In other words, simply adding real estate to a mixed-asset portfolio is not enough to ensure significant increases in performance as the results are dependent on the percentage added and the proper reallocation of the initial portfolio mix in the expanded portfolio.

Evaluating the effectiveness of common structures in property portfolio construction

A good portfolio structure enables an investor to diversify more effectively and understand systematic influences on their performance. However, in the property market, the choice of structure is affected by data constraints and convenience. Using individual return data, this study tests the hypothesis that some common structures in the UK do not explain a significant amount about property returns. It is found that, in the periods studied, not all the structures were effective and, for the annual returns, no structures were significant in all periods. The results suggest that the drivers represented by the structures take some time to be reflected in individual property returns. They also confirm the results of other studies in finding property type a much stronger factor in explaining returns than regions.