The first part of this paper gives an overview of a simplified approach to the statistical analysis of PET and fMRI data, including new developments and future directions. The second part outlines a new method, based on multivariate linear models (MLM), for characterising the response in PET and fMRI data, which overcomes some of the drawbacks of current methods such as SSM, SVD, PLS and CVA.

Abstract— Meningioma is one of the most frequent tumors and grows on the surface of the brain. This pushes the brain leading to stress changes in the brain causing it to shift from its region. On a broader scale, two methods of estimation of brain shift are used. One is the non-linear model and the other is linear model. Linear model is further optimized to produce better results. Better accura...

Here it is shown that the use of weighted l2-norms is a useful tool for the analysis of adaptive control loops based on non-linear models. We apply a simpliied feedback linearizing controller based on a linearly parameterized non-linear discrete-time input/output model with slowly time-varying parameters. The main assumption that is required in order to apply weighted l2-norms for analysis is t...

Why do we build models? There are two basic reasons: explanation or prediction (Ripley, 2004). Using large ensembles of models for prediction is commonplace, but is rarely used for explanation, where we typically choose one “best” model. When there are several equally good models, it is common sense to look at them too, but can the “bad” models tell us something as well? This paper describes ex...

Technological advances have led to a proliferation of structured big data that have matrix-valued covariates. We are specifically motivated to build predictive models for multi-subject neuroimaging data based on each subject's brain imaging scans. This is an ultra-high-dimensional problem that consists of a matrix of covariates (brain locations by time points) for each subject; few methods curr...

We describe procedures for creating efficient spectral representations for color. The representations generalize conventional tristimulus representations, which are based on the peripheral encoding by the human eye. We use low-dimensional linear models to approximate the spectral properties of surfaces and illuminants with respect to a collection of sensing devices. We choose the linear-model b...

This work presents a technique for identifying the parameters of a continuous process using Genetic Algorithms. The flexibility of this technique allows the parameters identification of high order linear models, and non-linear models. One of the advantages is that any type of input signal can be used. This feature is very useful when an industrial process cannot be stopped for specific identifi...

This paper deals with the design of a nonlinear observer for the inverted pendulum. The observer uses the standard structure of the linear observer with the linear model replaced by a nonlinear model. Nominal observer gains are determined from a linearized model. This model is also used to find a compromise between robustness and performance. The stability of the observer is then discussed and ...

In this article we consider the stochastic restricted ridge estimation in semipara-metric linear models when the covariates are measured with additive errors. The development of penalized corrected likelihood method in such model is the basis for derivation of ridge estimates. The asymptotic normality of the resulting estimates are established. Also, necessary and sufficient condition...

Usually the existence of influential observations is complicated by the presence of collinearity in linear measurement error models. However no method of influence measure available for the possible effect's that collinearity can have on the influence of an observation in such models. In this paper, a new type of ridge estimator based corrected likelihood function (REC) for linear measurement e...