Prediction error property of the Lasso Estimator and its generalization

The lasso procedure is an estimator-shrinkage and variable selection method. This paper shows that there always exists an interval of tuning parameter values such that the corresponding mean squared prediction error for the lasso estimator is smaller than for the ordinary least squares estimator. For an estimator satisfying some condition such as unbiasedness, the paper defines the corresponding generalized lasso estimator. Its mean squared prediction error is shown to be smaller than that of the estimator for values of the tuning parameter in some interval. This implies that all unbiased estimators are not admissible. Simulation results for five models support the theoretical results.

Does the choice of estimator matter when forecasting returns?

While the literature concerned with the predictability of stock returns is huge, surprisingly little is known when it comes to role of the choice of estimator of the predictive regression. Ideally, the choice of estimator should be rooted in the salient features of the data. In case of predictive regressions of returns there are at least three such features; (i) returns are heteroskedastic, (ii) predictors are persistent, and (iii) regression errors are correlated with predictor innovations. In this paper we examine if the accounting of these features in the estimation process has any bearing on our ability to forecast future returns. The results suggest that it does.

Estimator-based sliding mode control strategy design

This study addresses the design and properties of serial sliding mode control (SMC) systems for an induction servo motor drive to track periodic commands. It contains a SMC, an adaptive SMC (ASMC) and an estimator-based SMC (ESMC). The effectiveness of the proposed control systems is verifi ed by numerical simulations, and the superiority of the ESMC system is indicated in comparison with the SMC and ASMC systems.

Exact-estimator-based terminal sliding mode control system design

This paper investigates an estimator-based terminal sliding mode control system. An exact estimator is proposed to exactly estimate the unknown uncertainties in finite time. The output of the exact estimator is used to design a continuous chattering free terminal sliding mode control. The time taken for the closed-loop system to reach zero tracking error is proven to be finite. Experiment results are presented, using a real time digital-signal-processor (DSP) based electromagnetic levitation system to implement the control performance.

Genetic algorithm based PI controller tuning for induction motor drive with ANN flux estimator

This paper presents a Genetic Algorithm (GA) based fast speed response controller for poly-phase induction motor drive. Here the proportional and integral gains of PI controller are optimized by GA to achieve quick speed response. An adaptive Recurrent Neural Network (RNN) with Real Time Recurrent Learning (RTRL) algorithm is proposed to estimate rotor flux. An online tuning scheme to update the weight of RNN is presented to overcome stator resistance variation problem. This tuning scheme requires torque estimator to calculate the torque error. Space vector modulation (SVM) technique is used to produce the motor input voltage. Simulation tests have been performed to study the dynamic performances of the drive system for both the classical PI and the genetic algorithm based PI controllers.

Vector control of a position sensorless SPMSM drive with RNN based stator flux estimator

A position sensorless Surface Permanent Magnet Synchronous Motor (SPMSM) drive based on single layer Recurrent Neural Network (RNN) is presented in this paper. The motor equations are written in rotor fixed d-q reference frame. A PID controller is used to process the speed error to generate the reference torque current keeping the magnetizing current fixed. The RNN estimator is used to estimate flux components along the stator fixed stationary axes. The flux angle and the reference current phasor angle are used in vector rotator to generate the reference phase currents. Hysteresis current controller block controls the switching of the three phase inverter to apply voltage to the motor stator. Simulation studies on different operating conditions indicate the acceptability of the drive system. The proposed estimator can be used to accurately measure the motor fluxes and rotor angle over a wide speed range. The proposed control scheme is robust under load torque disturbances and motor parameter variations. It is also simple and low cost to implememnt in a practical environment

EEG data classification using wavelet features selected by Wilcoxon statistics

This paper introduces a method to classify EEG signals using features extracted by an integration of wavelet transform and the nonparametric Wilcoxon test. Orthogonal Haar wavelet coefficients are ranked based on the Wilcoxon test’s statistics. The most prominent discriminant wavelets are assembled to form a feature set that serves as inputs to the naïve Bayes classifier. Two benchmark datasets, named Ia and Ib, downloaded from the brain–computer interface (BCI) competition II are employed for the experiments. Classification performance is evaluated using accuracy, mutual information, Gini coefficient and F-measure. Widely used classifiers, including feedforward neural network, support vector machine, k-nearest neighbours, ensemble learning Adaboost and adaptive neuro-fuzzy inference system, are also implemented for comparisons. The proposed combination of Haar wavelet features and naïve Bayes classifier considerably dominates the competitive classification approaches and outperforms the best performance on the Ia and Ib datasets reported in the BCI competition II. Application of naïve Bayes also provides a low computational cost approach that promotes the implementation of a potential real-time BCI system.

Selected Ranked Set Sampling

This paper proposes a sampling procedure called selected ranked set sampling (SRSS), in which only selected observations from a ranked set sample (RSS) are measured. This paper describes the optimal linear estimation of location and scale parameters based on SRSS, and for some distributions it presents the required tables for optimal selections. For these distributions, the optimal SRSS estimators are compared with the other popular simple random sample (SRS) and RSS estimators. In every situation the estimators based on SRSS are found advantageous at least in some respect, compared to those obtained from SRS or RSS. The SRSS method with errors in ranking is also described. The relative precision of the estimator of the population mean is investigated for different degrees of correlations between the actual and erroneous ranking. The paper reports the minimum value of the correlation coefficient between the actual and the erroneous ranking required for achieving better precision with respect to the usual SRS estimator and with respect to the RSS estimator.

Robust extended Kalman filter based technique for location management in PCS networks

Provisioning of real-time multimedia sessions over wireless cellular network poses unique challenges due to frequent handoff and rerouting of a connection. For this reason, the wireless networks with cellular architecture require efficient user mobility estimation and prediction. This paper proposes using Robust Extended Kalman Filter as a location heading altitude estimator of mobile user for next cell prediction in order to improve the connection reliability and bandwidth efficiency of the underlying system. Through analysis we demonstrate that our algorithm reduces the system complexity (compared to existing approach using pattern matching and Kalman filter) as it requires only two base station measurements or only the measurement from the closest base station. Further, the technique is robust against system uncertainties due to inherent deterministic nature in the mobility model. Through simulation, we show the accuracy and simplicity in implementation of our prediction algorithm.

Identifying markov blankets using lasso estimation

Determining the causal relation among attributes in a domain is a key task in data mining and knowledge discovery. The Minimum Message Length (MML) principle has demonstrated its ability in discovering linear causal models from training data. To explore the ways to improve efficiency, this paper proposes a novel Markov Blanket identification algorithm based on the Lasso estimator. For each variable, this algorithm first generates a Lasso tree, which represents a pruned candidate set of possible feature sets. The Minimum Message Length principle is then employed to evaluate all those candidate feature sets, and the feature set with minimum message length is chosen as the Markov Blanket. Our experiment results show the ability of this algorithm. In addition, this algorithm can be used to prune the search space of causal discovery, and further reduce the computational cost of those score-based causal discovery algorithms.

The test-retest reliability of the child-initiated pretend play assessment

OBJECTIVE: The purpose of this study is to establish the test–retest reliability of the Child-Initiated Pretend Play Assessment (ChIPPA) (Stagnitti, 2002a; Stagnitti, Unsworth, & Rodger, 2000).

METHOD: The first author rated 38 preschool children ages 4 and 5 years (4 with developmental delay and 34 typically developing) on the ChIPPA. The ChIPPA employs conventional play materials and unstructured play materials to assess three qualities of a child's play ability: elaborateness of play action, ability to substitute objects during play, and the child's need to imitate the modelled actions of the examiner. The ChIPPA was administered twice, at a 2-week interval, to each participant.

RESULTS: Test–retest intraclass correlation coefficients (ICCs) (Type 2,1) calculated for each of the three elaborate play measures ranged from .73 to .84. A test–retest ICC of .56 was obtained for object substitution with unstructured play materials. The test–retest ICC obtained for the combined score for unstructured and conventional play materials was .57. Percentage agreement figures ranging from 63.2% to 84.2% were obtained on test–retest of the object substitution with conventional toys and imitated actions measures. There was no significant difference between test and retest scores for these measures based on a Wilcoxon Matched Pairs Signed-Ranks Test (Wilcoxon Sign Test).

CONCLUSION: Elaborate play scores, object substitution with conventional toys score, and imitation scores on the ChIPPA showed stability over time. Object substitution scores using unstructured materials were the least stable play measures and appeared to be related to the child's play themes. Since play is the primary occupation of children, it is essential that therapists have a reliable measure of play behavior. The test–retest reliability results from the ChIPPA provide evidence that this assessment produces a stable measure of play behavior that can then guide therapists when planning intervention strategies for children.

Robustness to non-normality of various tests for the one-sample location problem

This paper studies the effect of the normal distribution assumption on the power and size of the sign test, Wilcoxon's signed rank test and the t-test when used in one-sample location problems. Power functions for these tests under various skewness and kurtosis conditions are produced for several sample sizes from simulated data using the g-and-k distribution of MacGillivray and Cannon.

Node localization using mobile robots in delay-tolerant sensor networks

We present a novel scheme for node localization in a Delay-Tolerant Sensor Network (DTN). In a DTN, sensor devices are often organized in network clusters that may be mutually disconnected. Some mobile robots may be used to collect data from the network clusters. The key idea in our scheme is to use this robot to perform location estimation for the sensor nodes it passes based on the signal strength of the radio messages received from them. Thus, we eliminate the processing constraints of static sensor nodes and the need for static reference beacons. Our mathematical contribution is the use of a Robust Extended Kalman Filter (REKF)-based state estimator to solve the localization. Compared to the standard extended Kalman filter, REKF is computationally efficient and also more robust. Finally, we have implemented our localization scheme on a hybrid sensor network test bed and show that it can achieve node localization accuracy within 1m in a large indoor setting.

Speed control and policing in a cellular mobile network: speedNet

In this paper, we describe SpeedNet, a GSM network variant which resembles an ad hoc wireless mobile network where base stations keep track of the velocities of mobile users (cars). SpeedNet is intended to track mobile users and their speed passively for both speed policing and control of traffic. The speed of the vehicle is controlled in a speed critical zone by means of an electro-mechanical control system, suitably referred to as VVLS (Vehicular Velocity Limiting System). VVLS is mounted on the vehicle and responds to the command signals generated by the base station. It also determines the next base station to handoff, in order to improve the connection reliability and bandwidth efficiency of the underlying network. Robust Extended Kalman Filter (REKF) is used as a passive velocity estimator of the mobile user with the widely used proportional and integral controller speed control. We demonstrate through simulation and analysis that our prediction algorithm can successfully estimate the mobile user’s velocity with low system complexity as it requires two closest mobile base station measurements and also it is robust against system uncertainties due to the inherent deterministic nature in the mobility model.