Variance Estimation in Censored Quantile Regression via Induced Smoothing

Statistical inference in censored quantile regression is challenging, partly due to the unsmoothness of the quantile score function. A new procedure is developed to estimate the variance of the Bang and Tsiatis inverse-censoring-probability weighted estimator for censored quantile regression by employing the idea of induced smoothing. The proposed variance estimator is shown to be asymptotically consistent. In addition, a numerical study suggests that the proposed procedure performs well in finite samples, and it is computationally more efficient than the commonly used bootstrap method.     

Estimation of Expected Shortfall Using Quantile Regression: A Comparison Study

Computational Economics - Expected Shortfall ( $$mathrm {ES}$$ ) is one of the most heavily used measures of financial risk. It is defined as a scaled integral of the quantile of the...     

On a Modification of the Multistage Bidding Model with Continuous Bids and Asymmetric Information

This paper considers a modification of the multistage bidding model with continuous bids. Bidding takes place between two players for one unit of a risky asset (one stock). Player 1 knows the real price of the asset while Player 2 knows only the probabilities of high and low prices of the asset. At each stage of the bidding, players make real valued bids. The higher bid wins, and one unit of the risky asset is transacted to the winning player. The price of the transaction is a convex combination of the bids with a given coefficient. The optimal strategies of the players and the value of the n-stage game are found.     

基于自主学习与SCAD-Net正则化的回归模型

众多基因生物标志物选择方法常因研究样本较少而不能直接用于临床诊断.于是有学者提出整合不同基因表达数据同时保留生物信息完整性的方法.然而,由于存在批量效应,导致直接整合不同基因表达数据可能会增加新的系统误差.针对上述问题,提出一个融合自主学习与SCAD-Net正则化的分析框架.一方面,自主学习方法能够先从低噪声样本中学习出基础模型,然后再通过高噪声样本学习使得模型更加稳健,从而避免批量效应;另一方面,SCAD-Net正则化融合了基因表达数据与基因间的交互信息,可以实现更好的特征选择效果.不同情形下的模拟数据以及在乳腺癌细胞系数据集上的结果表明,基于自主学习与SCAD-Net正则化的回归模型在处理高维复杂网络数据集时具有更好的预测效果.     

Model Learning with Local Gaussian Process Regression

Precise models of robot inverse dynamics allow the design of significantly more accurate, energy-efficient and compliant robot control. However, in some cases the accuracy of rigid-body models does not suffice for sound control performance due to unmodeled nonlinearities arising from hydraulic cable dynamics, complex friction or actuator dynamics. In such cases, estimating the inverse dynamics model from measured data poses an interesting alternative. Nonparametric regression methods, such as Gaussian process regression (GPR) or locally weighted projection regression (LWPR), are not as restrictive as parametric models and, thus, offer a more flexible framework for approximating unknown nonlinearities. In this paper, we propose a local approximation to the standard GPR, called local GPR (LGP), for real-time model online learning by combining the strengths of both regression methods, i.e., the high accuracy of GPR and the fast speed of LWPR. The approach is shown to have competitive learning performance for high-dimensional data while being sufficiently fast for real-time learning. The effectiveness of LGP is exhibited by a comparison with the state-of-the-art regression techniques, such as GPR, LWPR and ν-support vector regression. The applicability of the proposed LGP method is demonstrated by real-time online learning of the inverse dynamics model for robot model-based control on a Barrett WAM robot arm.     

偏最小二乘回归模型的泛化能力改进研究

为了提高模型的泛化能力,提出了嵌入缩放思想的偏最小二乘回归(Partial Least-Squares Regression,PLS)建模方法。该方法通过对输入向量的缩放处理,将训练样本模糊化,缩小测试误差,从而提高了PLS的泛化能力。对原有的缩放法进行了改进,提出r算法。该算法可以找到合适的缩放因子,得到泛化能力更强的模型。仿真实验证明了所提方法的有效性。     

带约束的最优多元回归模型及其应用

针对区域经济发展短期预测建模的特点,结合核方法和支持向量回归的研究进展,提出了一类带约束的最优多元回归模型,该模型综合考虑了多元回归函数的拟合误差、泛化能力以及经济预测的特点,为区域经济的发展预测提供了一种新方案,对广东省江门市最近五年经济发展的预测也验证了该模型的有效性.     

Stochastic Model of Network Formation with Asymmetric Players

Automation and Remote Control - We propose a network formation model using the theory of stochastic games with random terminal time. Initially, the leader proposes a joint project in the form of a...     

Inferring Causal Interactions in Financial Markets Using Conditional Granger Causality Based on Quantile Regression

Computational Economics - Granger causality analysis emerges as a typical method for inferring causal interactions in economics variables. Yet the traditional pairwise approach to Granger causality...     

Forecasting financial time series volatility using Particle Swarm Optimization trained Quantile Regression Neural Network

Accurate forecasting of volatility from financial time series is paramount in financial decision making. This paper presents a novel, Particle Swarm Optimization (PSO)-trained Quantile Regression Neural Network namely PSOQRNN, to forecast volatility from financial time series. We compared the effectiveness of PSOQRNN with that of the traditional volatility forecasting models, i.e., Generalized Autoregressive Conditional Heteroskedasticity (GARCH) and three Artificial Neural Networks (ANNs) including Multi-Layer Perceptron (MLP), General Regression Neural Network (GRNN), Group Method of Data Handling (GMDH), Random Forest (RF) and two Quantile Regression (QR)-based hybrids including Quantile Regression Neural Network (QRNN) and Quantile Regression Random Forest (QRRF). The results indicate that the proposed PSOQRNN outperformed these models in terms of Mean Squared Error (MSE), on a majority of the eight financial time series including exchange rates of USD versus JPY, GBP, EUR and INR, Gold Price, Crude Oil Price, Standard and Poor 500 (S&P 500) Stock Index and NSE India Stock Index considered here. It was corroborated by the Diebold–Mariano test of statistical significance. It also performed well in terms of other important measures such as Directional Change Statistic (Dstat) and Theil's Inequality Coefficient. The superior performance of PSOQRNN can be attributed to the role played by PSO in obtaining the better solutions. Therefore, we conclude that the proposed PSOQRNN can be used as a viable alternative in forecasting volatility.     

Monte Carlo Method and Quantile Regression for Uncertainty Analysis of Wind Power Forecasting Based on Chaos-LS-SVM

International Journal of Control, Automation and Systems - In the paper, the chaos least squares support vector machine algorithm (Chaos-LS-SVM) is applied. To conduct uncertainty analysis of wind...     

A Regularized Contact Model with Asymmetric Damping and Dwell-Time Dependent Friction

A general regularized contact model, including normal compliance, energydissipation, and tangential friction, is described in this paper. Thenormal damping coefficient is formulated as a function of the coefficientof restitution e and the impact velocity only; the results areenergy-consistent, with continuous force progression at the beginningand end of the impact, for both small and large values of e.The introduced seven parameter friction model based on an explicitformulation of the friction forces is suitable for real-timeapplications. The friction forces are split into its sliding andsticking contribution and a temporal lag effect, the dwell-time, isincluded using a novel dwell-time dependent stick state variable. Several examples are presented to demonstrate the features of thisgeneral contact model. The simulation results for a double pendulumhitting a plane are obtained, and a comparison with a benchmark problem shows the model behavior is in good agreement with published results.     

Handling Model Changes: Regression Testing and Test-Suite Update with Model-Checkers

Several model-checker based methods to automated test-case generation have been proposed recently. The performance and applicability largely depends on the complexity of the model in use. For complex models, the costs of creating a full test-suite can be significant. If the model is changed, then in general the test-suite is completely regenerated. However, only a subset of a test-suite might be invalidated by a model change. Creating a full test-suite in such a case would therefore waste time by unnecessarily recreating valid test-cases. This paper investigates methods to reduce the effort of recreating test-suites after a model is changed. This is also related to regression testing, where the number of test-cases necessary after a change should be minimized. This paper presents and evaluates methods to identify obsolete test-cases, and to extend any given test-case generation approach based on model-checkers in order to create test-cases for test-suite update or regression testing.     

Quantile regression with doubly censored data

Quantile regression offers a semiparametric approach to modeling data with possible heterogeneity. It is particularly attractive for censored responses, where the conditional mean functions are unidentifiable without parametric assumptions on the distributions. A new algorithm is proposed to estimate the regression quantile process when the response variable is subject to double censoring. The algorithm distributes the probability mass of each censored point to its left or right appropriately, and iterates towards self-consistent solutions. Numerical results on simulated data and an unemployment duration study are given to demonstrate the merits of the proposed method.     

On Convergence of a Stochastic Quasigradient Algorithm of Quantile Optimization

For the nonantagonistic two-person game which is equivalent to the problem of minimizing the quantile function, a modification of the stochastic quasigradient algorithm to seek the Nash point was proposed. The Nash point defines both the optimal strategy minimizing the quantile function and the minimum value of this function. Convergence of the algorithm with the probability 1 was proved. The question of choosing the starting point was discussed.     

样本加权的多模型回归

回归分析是一种建立变量之间函数关系的简便方法.原始的回归分析算法并未考虑样本点的权重,即认为每个样本点的重要性都相等.但是,这样的算法在遇到包含野值点的实际问题时经常会失效,因为野值点会对回归模型产生很大的干扰.而对于多模型回归估计,每个样本点隶属于各模型的程度不同.针对多模型回归的这一特点,研究一种自适应的样本加权方法,在每一次计算样本点隶属度时,也对样本点的权重进行逼近,尽可能使野值点的权重减小为0,数值实验表明了该方法的有效性.     

Model Selection for Small Sample Regression

Model selection is an important ingredient of many machine learning algorithms, in particular when the sample size in small, in order to strike the right trade-off between overfitting and underfitting. Previous classical results for linear regression are based on an asymptotic analysis. We present a new penalization method for performing model selection for regression that is appropriate even for small samples. Our penalization is based on an accurate estimator of the ratio of the expected training error and the expected generalization error, in terms of the expected eigenvalues of the input covariance matrix.     

带隐变量的回归模型EM算法

带有隐变量的回归模型具有非常广泛的应用场合,隐回归模型的参数求解问题依赖于自变量的分布假设。基于自变量的beta分布的假设条件,给出了隐回归模型的EM算法,详细地推导了模型中的参数求解过程,给出了使用牛顿法求解beta分布参数的算法,并提出一个合适的初值选择算法。在模拟数据和真实数据的基础上进行了详细的比较性试验,结果表明,对具有不同分布特征的因变量观察值,EM算法能够有效地求解隐回归模型的参数。     

Asymmetric Adjustment and Bias in Estimation of an Equilibrium Relationship from a Cointegrating Regression

This paper uses Monte Carlo methods to investigate the effects of asymmetricadjustment on estimates of the parameters of the equilibrium relationshipbetween a set of variables. We demonstrate that simple least squares estimatesand the implicit estimates from a symmetric error correction model both leadto biases in the constant term. This bias increases with the size of theasymmetry and shows no tendency to decline with the sample size. We also showthat if the biased estimates of the equilibrium relationship are then used todevide the sample into different regimes to test for assymmetric adjustment,then the resulting test has low power. The power of tests for asymmetry canbe increased significantly by using simultaneous estimation of the parametersof the equilibrium relationship and the asymmetric adjustment process.     

多视角级联回归模型人脸特征点定位

针对人脸姿态偏转较大导致人脸特征点定位精度低的问题，提出了多视角人脸特征点定位算法，采用随机森林局部学习与全局线性回归相结合的级联姿态回归（Cascaded Pose Regression，CPR）人脸特征点定位模型，在不同的人脸姿态视角下建立不同的模型，以多模型代替单一模型来提高人脸特征点定位的精度。首先采用CPR模型对不同视角下的人脸建立不同的模型；然后采用多视角生成模型（Multi-View Generative Model，MVGM）来评估输入人脸图片的姿态；最后根据评估的姿态选择相对应的模型，进而实现特征点的精确定位。仿真实验结果表明，相比于现有的几种人脸特征点定位算法，所提算法实现了更精确的定位效果。