COVARIATE-ADJUSTED REGRESSION FOR LONGITUDINAL DATA INCORPORATING CORRELATION BETWEEN REPEATED MEASUREMENTS

We propose an estimation method that incorporates the correlation/covariance structure between repeated measurements in covariate-adjusted regression models for distorted longitudinal data. In this distorted data setting, neither the longitudinal response nor (possibly time-varying) predictors are directly observable. The unobserved response and predictors are assumed to be distorted/contaminated by unknown functions of a common observable confounder. The proposed estimation methodology adjusts for the distortion effects both in estimation of the covariance structure and in the regression parameters using generalized least squares. The finite-sample performance of the proposed estimators is studied numerically by means of simulations. The consistency and convergence rates of the proposed estimators are also established. The proposed method is illustrated with an application to data from a longitudinal study of cognitive and social development in children.     

The effect of tapering on the semiparametric estimators for nonstationary long memory processes

In this paper, we study, by a Monte Carlo simulation, the effect of the order p of “Zhurbenko-Kolmogorov” taper on the asymptotic properties of semiparametric estimators. We show that p  =  [d + 1/2] + 1 gives the smallest variances and mean squared errors. These properties depend also on the truncation parameter m. Moreover, we study the impact of the short-memory components on the bias and variances of these estimators. We finally carry out an empirical application by using four monthly seasonally adjusted logarithm Consumer Price Index series.      

CANONICAL VARIATE ANALYSIS WITH SPATIALLY-CORRELATED DATA

A general canonical variate model is derived when the observations are spatially correlated. For spatial covariance structures resulting from dependence of a pixel on its nearest neighbours, the solution reduces to an analysis of neighbour-corrected values. The usual analysis, in which spatial correlation is ignored, gives similar canonical vectors but over-estimates the canonical roots. A formula for approximating the reduction in the canonical roots to adjust for the spatial correlation is given.     

Portfolio optimization: New capabilities and future methods

Summary  This paper addresses the problem of portfolio selection in finance. In many cases, currently available software to compute the efficient frontier runs into difficulty in problems with more than about 600 securities. To proceed beyond this size, it is often necessary to modify the problem in which case there is typically a loss of information. In this paper, we discuss a computer capability that can exactly compute mean-variance efficient frontiers of problems with up to 2,000 securities in very reasonable time (even if a problem’s covariance matrix is 100% dense). The paper also discusses an augmentation to the theory of portfolio selection that allows multiple objectives (such as dividends, liquidity, social responsibility, amount invested in R&D, and so forth) to be incorporated into the portfolio selection process. In such problems, the efficient set is no longer a “frontier,” but is now best described as a “surface” with the interesting property that it is composed of platelets (like on the back of a turtle). Moreover, the computer capability that can compute the exact efficient frontier of a mean-variance problem with up to 2,000 securities also has, after additional coding, the ability to compute exactly all platelets of the efficient surface of a tri-criterion portfolio problem with up to 400 securities.

---

Zusammenfassung  In dieser Arbeit stellen wir einen leistungsf?higen Rechenalgorithmus vor, um den effizienten Rand (die nichtdominierten Alternativen) von Portfolio-Auswahlproblemen in der Finanzierung zu bestimmen. Wir bezeichnen den Berechnungsalgorithmus, der in Java programmiert ist, mit MPQ (multi-parametric quadratic programming). MPQ weist gegenüber bisherigen Berechnungsverfahren eine Reihe von Vorteilen auf: Es kann für umfangreiche Anwendungsf?lle genutzt werden, ist durch passable Rechenzeiten charakterisiert und kann die Menge effizienter Alternativen in einem Bruchteil bisher üblicher Rechenzeiten bestimmen.

     

证券组合投资收益和风险的量化分析

现代理财学的重要内容之一是投资管理。投资管理最重要的环节是把握投资机会和明智地进行投资决策。其目标是追求利润最大化 ,风险最小化。风险与收益是现代理财学中的一对基本概念 ,必须正确地处理风险与收益这一对理财过程中的基本矛盾。在追求利润最大化的同时 ,进行多元化投资 ,以分散投资风险     

Varying-time random effects models for longitudinal data: unmixing and temporal interpolation of remote-sensing data

Remote sensing is a helpful tool for crop monitoring or vegetation-growth estimation at a country or regional scale. However, satellite images generally have to cope with a compromise between the time frequency of observations and their resolution (i.e. pixel size). When concerned with high temporal resolution, we have to work with information on the basis of kilometric pixels, named mixed pixels, that represent aggregated responses of multiple land cover. Disaggreggation or unmixing is then necessary to downscale from the square kilometer to the local dynamic of each theme (crop, wood, meadows, etc.).

Assuming the land use is known, that is to say the proportion of each theme within each mixed pixel, we propose to address the downscaling issue through the generalization of varying-time regression models for longitudinal data and/or functional data by introducing random individual effects. The estimators are built by expanding the mixed pixels trajectories with B-splines functions and maximizing the log-likelihood with a backfitting-ECME algorithm. A BLUP formula allows then to get the ‘best possible’ estimations of the local temporal responses of each crop when observing mixed pixels trajectories. We show that this model has many potential applications in remote sensing, and an interesting one consists of coupling high and low spatial resolution images in order to perform temporal interpolation of high spatial resolution images (20 m), increasing the knowledge on particular crops in very precise locations.

The unmixing and temporal high-resolution interpolation approaches are illustrated on remote-sensing data obtained on the South-Western France during the year 2002.     

Modified minimum covariance determinant estimator and its application to outlier detection of chemical process data

To overcome the main flaw of minimum covariance determinant (MCD) estimator, i.e. difficulty to determine its main parameter h, a modified-MCD (M-MCD) algorithm is proposed. In M-MCD, the self-adaptive iteration is proposed to minimize the deflection between the standard deviation of robust mahalanobis distance square, which is calculated by MCD with the parameter h based on the sample, and the standard deviation of theoretical mahalanobis distance square by adjusting the parameter h of MCD. Thus, the optimal parameter h of M-MCD is determined when the minimum deflection is obtained. The results of convergence analysis demonstrate that M-MCD has good convergence property. Further, M-MCD and MCD were applied to detect outliers for two typical data and chemical process data, respectively. The results show that M-MCD can get the optimal parameter h by using the self-adaptive iteration and thus its performances of outlier detection are better than MCD.     

An improved meta-analysis for analyzing cylindrical-type time series data with applications to forecasting problem in environmental study

In this paper, we propose an improved generalized least square (GLS) meta-analysis in a linear-circular regression, and show its utility in the analysis of a certain environmental issue. The existing GLS meta-analysis proposed in Becker and Wu has a serious flaw since information about the covariance among coefficients across studies is not utilized. In our proposed meta-analysis, we take the correlations between adjacent studies into account, and improve the existing GLS meta-analysis. We provide numerical examples to compare the proposed method with several other existing methods by using Akaike's Information Criterion, Bayesian Information Criterion and mean square prediction errors with applications to forecasting problem in Environmental study.     

The use of multilevel modeling and the level two residual file to explore the relationship between middle years programme student performance and diploma programme student performance

Multilevel modeling has recently found a substantial niche in the context of educational research, although several details about the methodological application of these models have yet to be explored in an achievement data framework. This paper makes use of data provided by the International Baccalaureate (IB) in order to investigate modeling decisions and certain applications of the level two residual file in an effort to increase understanding about the way linear and logistic multilevel models function. The focus of this research is on the relationship between performances in two IB programmes: the Middle Years Programme (MYP) and the Diploma Programme (DP). The impact of predictors on the interpretation of the unconditional and conditional variance-covariance matrix as well as the reliability coefficients is discussed. Empirical findings suggest that students who perform better during MYP moderation tend to perform better on DP exams.