We propose a computer intensive method for linear dimension reduction which minimizes the classification error directly. Simulated annealing (Bohachevsky et al 1986) as a modern optimization technique is used to solve this problem effectively. This approach easily allows to incorporate user...

We propose multivariate classification as a statistical tool to describe business cycles. These cycles are often analyzed as a univariate phenomenon in terms of GNP or industrial net production ignoring additional information in other economic variables. Multivariate classification overcomes...

The best linear unbiased estimator BLUE (CXb) of a linear transform CX b in the general Gauss-Markov model (y, E (y) = X b Cov (y) =a2v) is the linear transform C BLUE (Xb) of the best linear unbiased estimator BLUE (Xb) of Xb. Similarly, for the ordinary least squares estimator OLSE (CXb) = C...

Support Vector Machines (SVMs) have become a popular tool for learning with large amounts of high dimensional data. However, it may sometimes be preferable to learn incrementally from previousSVM results, as computing a SVM is very costly in terms of time and memory consumption or because the...

We show that the weak Pareto law, as used to characterize the tail behaviour of income distributions, implies regularly varying tail probabilities, but that the reverse implication does not hold. We also establish implications among other versions of the weak Pareto law.

Das Abflussverhalten des Rheins wird mittels flexibler saisonaler Modelle mit langem Gedächtnis modelliert. Zur Schätzung der Persistenz wird für jede Saisonfrequenz separat eine Log-Periodogramm Regression durchgeführt. Verglichen mit Standard-ARMA-Prozessen liefern diese Modelle eine gute...

An unbiased point estimator T for an unknown parameter can be improved in the sense of the Mean Squared Error (MSE) by T T for suitable factors . Here, we want to discuss this approach in the context of combination of forecasts. We consider the shrinkage technique for...

The paper presents an approach to the analysis of data that contains (multiple) structural changes in a linear regression setup. We implement various strategies which have been suggested in the literature for testing against structural changes as well as a dynamic programming algorithm for the...

Time series analysis is an important and complex problem in machine learning and statistics. Real-world applications can consist of very large and high dimensional time series data. Support Vector Machines (SVMs) are a popular tool for the analysis of such data sets. This paper presents some SVM...