Discrimination between alternative arima model specifications

Summary The use of ARIMA models in applied economic forecasting has become more and more popular in recent years. In order to secure a good forecasting performance of these models, great care should be spent on the correct identification and the efficient estimation of models of this type.The present paper deals with problems often encountered in applied time series analysis. When fitting an ARIMA model to an observed macroeconomic time series, it might be detected that several models seem to represent this series equally well. The problem here is how to discriminate between alternative ARIMA model specifications representing the same macroeconomic time series.In the first part of the paperthe theoretical concepts, used in the identification and estimation of an ARIMA model, are briefly reviewed. In the second part, these techniques are then applied in a case-study in order to obtain ARIMA models for private consumer expenditure and disposable personal income.

---

Zusammenfassung Die vorliegende Arbeit beschäftigt sich mit einigen Problemen der angewandten Zeitreihen-analyse. Beim Versuch, ein ARIMA-Modell für beobachtete makroökonomische Zeitreihen zu schätzen, entdeckt man bisweilen, daß mehrere Modelle die Zeitreihe gleich gut zu repräsentieren scheinen. Hier stellt sich dann die Frage, ob es verläßliche Kriterien gibt, die eine Auswahl aus alternativen Modellansätzen gestatten. Im ersten Teil der Arbeit werden kurz die Konzepte, die für die Identifikation und Schätzung von ARIMA-Modellen zur Verfügung stehen, dargestellt. Der zweite Teil stellt eine Fallstudie dar, in der ARIMA-Modelle für den privaten Konsum und das verfügbare persönliche Einkommen entwickelt werden.

---

---

I would like to thankJohannes Ledolter from the University of Iowa for his very helpful comments on an earlier draft of this paper.     

Stabilitätskonzepte am Beispiel Österreichs

Summary This paper examines two aspects of equilibrium — and stability concepts, which go beyond the traditional analyses. Firstly the speed by which a systems returns to the equilibrium position after an initial disturbance and, secondly, the introduction of stochastic elements in the analysis of stability. Both aspects are presented against the background of a simple model of the Austrian economy.It is known that a system of difference equations is stable if the greatest sigenvalue is smaller than one. In trying to estimate confidence intervals it can be shown that an evaluation according to deterministic criteria alone can indeed be misleading, even though the probability of an instable solution is small in the model under examiniation.The speed, by which a system returns to the equilibrium position after an initial disturbance, is measured by the half-life-period. This is the time it takes for the effects of the initial disturbance to be reduced to half their original values. In the model under examination the half-line amounts to one year (real solution) or two years (complex solution) respectively. If we take the stochastic nature of the eigenvalues into consideration, we get, for the 95% level of significance, figures of 3 months to 6 years in the real case. The large spread results from the fact that the half-life-concept is very sensitive even towards minor changes in the eigenvalues.This paper shows that the conventional methods of estimation of coefficients in stability-analyses are not sufficiently accurate.     

Editorial Announcement

At the     

How Political Institutions Determine Corporate Governance Reforms: The Polity,Law and Corporate Practices in the Case of Switzerland

The aim of this article is to contribute to our understanding of the role that political institutions play in the reform processes of national corporate governance regimes. I argue that existing theories are limited in that they conceive of political institutions' impact on corporate governance largely in terms of a deterministic influence on the political coalitions that emerge and on the policy outcomes. Based on the analysis of the paradigmatic case of a consensual democracy – Switzerland – I show that the consensual polity does not directly determine the outcome of a reform process, but rather the direction of causality between legal changes and changes in practices as well as the nature of the changes. Consensual polities require large parliamentary majorities for legal change to happen. Therefore, corporate practices are likely to change before legal rules and less demanding forms of institutional change, such as ‘layering’, are favoured.     

Hierarchical Models for the Analysis of Likert Scales in Regression and Item Response Analysis

Appropriate modelling of Likert‐type items should account for the scale level and the specific role of the neutral middle category, which is present in most Likert‐type items that are in common use. Powerful hierarchical models that account for both aspects are proposed. To avoid biased estimates, the models separate the neutral category when modelling the effects of explanatory variables on the outcome. The main model that is propagated uses binary response models as building blocks in a hierarchical way. It has the advantage that it can be easily extended to include response style effects and non‐linear smooth effects of explanatory variables. By simple transformation of the data, available software for binary response variables can be used to fit the model. The proposed hierarchical model can be used to investigate the effects of covariates on single Likert‐type items and also for the analysis of a combination of items. For both cases, estimation tools are provided. The usefulness of the approach is illustrated by applying the methodology to a large data set.     

Competing risks models in discrete time with nominal or ordinal categories of response

Two principle approaches to the modelling of competing risks in discrete time are considered. In the first approach which is based on the separation between failure and cause specific response only the causes of failure are considered as ordered. The second approach which is based on the conditional response given interval [a _t-1,a_r) is reached allows for an ordering of causes of failureand the category ‘no failure’. The latter approach is shown to be more general. It is shown that the considered competing risks models may be estimated within the framework of generalized linear models. A data set concerning duration of unemployment illustrates the approaches.     

Merkmale einer vertrauensvollen Kommunikation zwischen Organisationen – eine empirische Analyse

In the business world one of the key drivers to reduce uncertainty is trust. It is based on an empirical study (Kano-analysis, 60% return rate) and identifies relevant factors of a trustful communication in long-term business relationships between organizations. Three types of factors can be identified: Must-be factors are a prerequisite of any relation and have a fundamental character in long-term relationships. The customers expect such factors to be an unquestionable part of any business relation (e.g. honesty). While attractive factors always boost the stability of trust in the relation (e.g. sympathy), one-dimensional factors not only stabilize trust if implemented, but can also deteriorate the relation if ignored (e.g. timeliness).     

Identifying structural VARs from sparse narrative instruments: Dynamic effects of US macroprudential policies

We study identification in Bayesian proxy VARs for instruments that consist of sparse qualitative observations indicating the signs of shocks in specific periods. We propose the Fisher discriminant regression and a non-parametric sign concordance criterion as two alternative methods for achieving correct inference in this case. The former represents a minor deviation from a standard proxy VAR, whereas the non-parametric approach builds on set identification. Our application to US macroprudential policies finds persistent declines in credit volumes and house prices together with moderate declines in GDP and inflation and a widening of corporate bond spreads after a tightening of capital requirements or mortgage underwriting standards.     

Sold Below Value? Why Takeover Offers Can Have Negative Premiums

Many studies have acknowledged the existence of negative offer premiums where the initial bid undercuts the target's preannouncement market price. However, this phenomenon has not been explained. Negative premiums occur frequently and are no measurement error. We demonstrate theoretically and empirically that “hidden earnouts,” where target shareholders participate in the bidder's share of joint synergies, and corrections of overvaluation explain negative premiums. We find that target shareholders profit from the consummation of a takeover even if the announced offer has a negative premium. Our theory generalizes to low positive premiums with predictive power for the bottom 25% of all premiums.