Model averaging with ridge regularization

Least-squares forecast averaging

Zusammenfassung Erstmals enthält der faktisch anonymisierte Mikrozensus (FAMZ) 1996 Stichprobeninformationen, die eine Berechnung der Varianz von Populationsschätzern ermöglichen. Nach der Darstellung der Ziehung des Mikrozensus (MZ) und des FAMZ wird ein methodisches Konzept zur Berechnung der Varianz entwickelt und diskutiert, das mit Standard-Software umgesetzt werden kann. Dieses Konzept wird auf die Schätzung von Totals, Verhältniszahlen und Anteilen angewendet. Neu ist die Behandlung von Hochrechnungsergebnissen nach der Anpassung an die Bevölkerungsfortschreibung im Rahmen eines Regressionsansatzes. Schließlich wird überprüft, inwieweit die Design-Zuschlagsfaktoren, das bisher einzige Instrument der FAMZ-Nutzer zur Schätzung der Varianz, brauchbare Ergebnisse auch für den FAMZ liefern. Im empirischen Teil werden für angewählte Merkmale Varianzschätzungen von MZ und FAMZ miteinander verglichen. Es wird gezeigt, dass die Varianz-Komponente bezüglich der zweiten Auswahlstufe, der Substichprobenziehung des FAMZ aus dem MZ, vernachlässigt werden kann, was die Berechnung der Varianzen erheblich vereinfacht. Ein weiteres Ergebnis besagt, dass die Varianzvergrößerung des FAMZ gegenüber dem MZ geringer ausfällt als dies bei einer einfachen Stichprobenziehung zu erwarten ist. Schließlich wird gezeigt, dass die Approximation der Varianz über die Design-Zuschlagsfaktoren auch für den FAMZ zu brauchbaren Ergebnissen führt, in Einzelfällen aber mit erheblichen Über- bzw. Unterschätzungen der Varianz verbunden ist.

This article examines high-dimensional covariates in regression discontinuity design (RDD) analysis. We introduce estimation and inference methods for the RDD models that incorporate covariate selection while maintaining stability across various numbers of covariates. The proposed methods combine a localization approach using kernel weights with $\ell _{1}$ -penalization to handle high-dimensional covariates. We provide both theoretical and numerical evidence demonstrating the efficacy of our methods. Theoretically, we present risk and coverage properties for our point estimation and inference methods. Conditions are given under which the proposed estimator becomes more efficient than the conventional covariate adjusted estimator at the cost of an additional sparsity condition. Numerically, our simulation experiments and empirical examples show the robust behaviors of the proposed methods to the number of covariates in terms of bias and variance for point estimation and coverage probability and interval length for inference.

Growth, Inequality and Poverty Relationships

It is well known that the variances of the least squares estimates are large and they can be far away from their true values in the case of multicollinearity. Therefore, the ridge regression method can be used as an alternative to the least squares method. However, the ridge estimator has a disadvantage that its distribution is unknown, so only asymptotic confidence intervals are obtained. The purpose of this paper is to study the impact of several ridge regularization parameters on the mean interval lengths of the confidence intervals and coverage probabilities constructed by the ridge estimator. A bootstrap method for the selection of ridge regularization parameter is used as well as the parametric methods. In order to compare the confidence intervals, standard normal approximation, student-t approximation and bootstrap methods are used and comparison is illustrated via real data and simulation study. The simulation study shows that the bootstrap choice of ridge regularization parameter yields narrower standard normal approximated confidence intervals than the PRESS choice of ridge regularization parameter but wider standard normal approximated, student-t approximated and bootstrap confidence intervals than the GCV choice of ridge regularization parameter.

The profile of time allocation in the metabolic pattern of society: An internal biophysical limit to economic growth

The purpose of this study is to explore the appropriate role for government when the real estate market fails to work efficiently, and to discuss based on previous experiences of Taiwan how effectiveness of the governmentís intervention was. When there is a market failure, many will expect the government intervention. However they often ignore the accompanying intervention costs that may induce government failure. The paper first reviews the theoretical rationale for government intervention in a market failure and then proposes a conceptual structure, based on the New Institutional Economics theory, of the appropriate role of government in the real estate market to minimize the costs generated from market failure and government failure. The proposed theory is verified by the empirical example of Taiwanís real estate market experiences during the three abnormal price fluctuations in 1974,1980, and 1988. The methods and extents of government interventions in these three crises are examined to understand whether they hinder or facilitate the efficient allocation of resources in the real estate market and whether the interventions promote the economic growth. Finally a summary and conclusion of this study is given.