Klaus Abberger

Quantile smoothing in financial time series

Various parametric models have been designed to analyze volatility in time series of financial market data. For maximum likelihood estimation these parametric methods require the assumption of a known conditional distribution. In this paper we examine the conditional distribution of daily DAX returns with the help of nonparametric methods. We use kernel estimators for conditional quantiles resulting from a kernel estimation of conditional distributions.

A simple graphical method to explore tail-dependence in stock-return pairs

For a bivariate data set the dependence structure cannot only be measured globally, for example with the Bravais–Pearson correlation coefficient, but the dependence structure can also be analysed locally. In this article the exploration of dependencies in the tails of the bivariate distribution is discussed. For this a graphical method which is called a chi-plot and which was introduced by Fisher and Switzer is used. Examples with simulated data sets illustrate that the chi-plot is suitable for the exploration of dependencies. This graphical method is then used to examine stock-return pairs. The kind of tail-dependence between returns has consequences, for example, for the calculation of the value at risk and should be modelled carefully. The application of the chi-plot to various daily stock-return pairs shows that different dependence structures can be found. This graph can therefore be an interesting aid for the modelling of returns.

Mikrodaten im ifo Institut für Wirtschaftsforschung – Bestand, Verwendung und Zugang

ZusammenfassungDas ifo Institut für Wirtschaftsforschung in München führt seit seiner Gründung im Jahr 1949 regelmäßige Unternehmensbefragungen durch. Bisher bestand kein institutionalisierter Datenzugang zu den Mikrodaten für Forscher außerhalb des Institutes. Die Mikrodaten werden nunmehr im ifo Datapool zusammengefasst. Dieser enthält neben externen Daten umfangreiche Datensätze der vier Standardumfragen des ifo Institutes: des Konjunkturtests, des Investitionstests, des Innovationstests und des World Economic Survey. In diesem Beitrag werden Umfang, Bestand, Dokumentation und Zugang zum neuen ifo Datapool beschrieben. AbstractThe Ifo Institute for Economic Research has been conducting regular business surveys since its foundation in 1949. Until now there was no institutionalized data access to the micro data for external researchers. The micro data are now being stored in the ifo Datapool. Besides external data, the ifo Datapool contains rich data sets of the four standard surveys of the institute: the business cycle test, the investment test, the innovation test, and the World Economic Survey. In this paper the coverage, documentation and access to the ifo Datapool is described.

The KOF Economic Barometer, Version 2014: A Composite Leading Indicator for the Swiss Business Cycle

This paper presents a composite leading indicator for the Swiss business cycle corresponding to the growth rate cycle concept. It is the result of a complete overhaul of the KOF Economic Barometer that has been published by the KOF Swiss Economic Institute on a monthly basis since 1976. In line with this tradition, the calculation of the new KOF Barometer comprises two main stages. The first consists of the variable selection procedure; and in the second stage these variables are subsequently transformed into one leading indicator. Whereas in the previous versions of the KOF Barometer six to 25 variables survived the first stage, the new – less discretionary and more automated – version of the first stage is much more generous. Currently, out of a set of 476 variables resulting in 4356 transformations thereof that are tested in the first stage, 219 variables manage to enter the second stage. The increased number of variables underlying the second stage allows a relatively stable and robust KOF Barometer – compared to its previous versions – that has hence no longer to rely on filtering techniques to reduce the noise in the final indicator. In a (pseudo-) real-time analysis the characteristics of the new KOF Barometer are compared to the previous versions and other alternatives.

Nonparametric Quantile Regression With Applications to Financial Time Series

For a random variable Y with covariate (vector) X and conditional cumulative distribution function F(y∣X = x) various approaches for non-parametric estimation of a conditional quantile function qα(x) = inf{y ∈ ℝF(y∣x) ≥ α} for 0 < α < 1 have been proposed in the past. They are either based upon an estimation of the conditional c.d.f. from which then an estimator for q α(x) is derived or on a local polynomial (mostly linear) approximation of q α(x) itself. Various variates of these two basic ideas were also discussed. Whereas an abundance of proposals exists for bandwidth selection in nonparametric regression, there is only little on smoothing parameter choice in quantile regression. Several possibilities, based on cross validation, generalized cross validation, plug in and double smoothing ideas are discussed in the paper. In a time series application, the covariate may consist of past time series values, leading to nonparametric quantile autoregression, or it may be the time index itself. In the latter case we arrive at quantile smoothing. Some applications to financial market time series are presented and compared with the outcomes of parametric GARCH models. Quantile forecasting based upon local quantile autoregression seems to be relatively new.

Is the Global Economy on the Brink of Recession

In 1937, in the midst of the US recovery from the Great Depression, President Roosevelt implemented spending cuts in pursuit of a balanced budget. Subsequently, the unemployment rate jumped nearly 6 percentage points over the next year and the US economy re-entered a major recession. In the midst of the current global recovery from the Great Recession, European and American policymakers again seem intent on pursuing the path of budget austerity. In light of slowing economic growth rates, shrinking consumer and business confi dence, and stubbornly high unemployment fi gures, could it be that Western economies are ignoring lessons from the past? DOI: 10.1007/s10272-011-0395-2

The effects of policy uncertainty on investment: Evidence from the unexpected acceptance of a far-reaching referendum in Switzerland

Does increased economic policy uncertainty dampen investment of firms? We provide direct evidence on this question by examining the effects of an unexpectedly accepted and far-reaching referendum in Switzerland in February 2014. The outcome put several economically relevant agreements between Switzerland and its main trading partner, the European Union, at stake. Using firm-level panel data covering the 2009–2015 period and data from two special surveys levied shortly after the vote, we examine how the induced policy uncertainty affected investment of Swiss firms. Our Difference-in-Differences-in-Differences estimations and complementary survey results give strong evidence that the uncertainty caused by the vote dampened, as theoretically expected, investment of exposed firms with irreversible investment by as much as one quarter in the two years following the vote. Exposed firms that can reverse investment appear rather to increase investment in the year after the vote. Our survey evidence suggests that these firms engage in investment to streamline their production.

Findings of the Signal Approach: A Case Study for Kazakhstan

This study concentrates on the signal approach for the monitoring of currency crises risks. It focuses on the properties of individual indicators prior to observed currency crises in Kazakhstan. The indicators are used to build composite indicators. An advanced approach uses principal components analysis for the construction of composite indicators. Furthermore, the common signal approach is improved by robust statistical methods. The estimation period reaches from 1997 to 2007. It is shown that most of the composite indicators are able to flag the reported crises in this time span at an early stage. In a second step it is checked whether the crisis observed in 2009 is signalled in advance.

Konjunkturanalyse: Prognose 2016/2017, Preissenkungen haben Einbruch abgefedert – Abwertung hellt Aussichten auf

This text contains the autumn forecast 2015 of the KOF Swiss Economic Institute at ETH Zurich, released on 1 October 2015. The first part discusses recent economic developments abroad and in Switzerland, and presents the main forecast results across the various sectors of the economy. In the second part, detailed forecasts for Switzerland follow, split up into the main demand components of GDP. We expect the Swiss economy to grow by 0.9% in 2015, 1.4% in 2016 and 1.8% in 2017, respectively, representing a small upward revision compared with our last forecast from June 2015. In particular, the positive economic development in Europe and the depreciation of the Swiss franc during the second half of 2015 are specified as reasons for this adjustment. The negative price developments during the first half of 2015 are also newsworthy, as both the general price level and sales prices in certain sectors are falling.

Ifo Survey on Employment Plans: Sectoral Evaluation

This Handbook aims to provide an overview of regular survey activities, as well as to show how survey results can be used scientifically in the context of business-cycle analysis and forecasting.