Forecasting and Structural Analysis with Contemporaneous Aggregates of Time Series Data

Description

Time series data in economics and social sciences are often constructed by aggregating over cross-sectional units such as individuals, sectors, states or countries. This type of aggregation is referred to as contemporaneous aggregation. Examples that highlight the practical importance include the sectoral aggregation in the construction of quarterly national accounts data, aggregation of different price index subcomponents, the aggregation of micro-based macroeconomic models and the construction of area-wide time series e.g. for the Euro area. Contemporaneous aggregation may change the dynamic properties (e.g. the persistence) of the time series data in a substantial way, such that often more complex time series models are needed for the aggregate. In this project we explore the consequences of contemporaneous aggregation for econometric forecasting and structural analysis. We focus on issues that arise (1) when there are structural changes in the data generating processes of some cross-sectional units and (2) when the time series information is missing for some cross-sectional units. These problems are of immense practical relevance and are likely to bias the information contained in the aggregate series and any econometric analysis based on this data. Alternative aggregation methods and alternatives to aggregation are developed and their usefulness in econometric forecasting and policy analysis models for aggregates is analyzed.
2.

Institutions
  • FB Wirtschaftswissenschaften
Publications
    Balabanova, Zlatina; Brüggemann, Ralf(2017): External Information and Monetary Policy Transmission in New EU Member States : Results from FAVAR Models Macroeconomic Dynamics ; 21 (2017), 02. - S. 311-335. - ISSN 1365-1005. - eISSN 1469-8056

External Information and Monetary Policy Transmission in New EU Member States : Results from FAVAR Models

×
Balabanova, Zlatina; Brüggemann, Ralf

Origin (projects)

  Zeng, Jing(2017): Forecasting Aggregates with Disaggregate Variables : Does Boosting Help to Select the Most Relevant Predictors? Journal of Forecasting ; 36 (2017), 1. - S. 74-90. - ISSN 0277-6693. - eISSN 1099-131X

Forecasting Aggregates with Disaggregate Variables : Does Boosting Help to Select the Most Relevant Predictors?

×
Including disaggregate variables or using information extracted from the disaggregate variables into a forecasting model for an economic aggregate may improve the forecasting accuracy. In this paper we suggest to use the boosting method to select the disaggregate variables which are most helpful in predicting an aggregate of interest. We conduct a simulation study to investigate the variable selection ability of this method. To assess the forecasting performance a recursive pseudo-out-of-sample forecasting experiment for six key Euro area macroeconomic variables is conducted. The results suggest that using boosting to select relevant predictors is a feasible and competitive approach in forecasting an aggregate.

Origin (projects)

  Brüggemann, Ralf; Jentsch, Carsten; Trenkler, Carsten(2016): Inference in VARs with conditional heteroskedasticity of unknown form Journal of Econometrics ; 191 (2016), 1. - S. 69-85. - ISSN 0304-4076. - eISSN 1872-6895

Inference in VARs with conditional heteroskedasticity of unknown form

×
We consider a framework for asymptotically valid inference in stable vector autoregressive (VAR) models with conditional heteroskedasticity of unknown form. A joint central limit theorem for the LS estimators of both the VAR slope parameters as well as the unconditional innovation variance parameters is obtained from a weak vector autoregressive moving average model set-up recently proposed in the literature. Our results are important for correct inference on VAR statistics that depend both on the VAR slope and the variance parameters as e.g. in structural impulse responses. We also show that wild and pairwise bootstrap schemes fail in the presence of conditional heteroskedasticity if inference on (functions) of the unconditional variance parameters is of interest because they do not correctly replicate the relevant fourth moments’ structure of the innovations. In contrast, the residual-based moving block bootstrap results in asymptotically valid inference. We illustrate the practical implications of our theoretical results by providing simulation evidence on the finite sample properties of different inference methods for impulse response coefficients. Our results point out that estimation uncertainty may increase dramatically in the presence of conditional heteroskedasticity. Moreover, most inference methods are likely to understate the true estimation uncertainty substantially in finite samples.

Origin (projects)

  Brüggemann, Ralf; Zeng, Jing(2015): Forecasting Euro-Area Macroeconomic Variables Using a Factor Model Approach for Backdating Oxford Bulletin of Economics and Statistics ; 77 (2015), 1. - S. 22-39. - ISSN 0305-9049. - eISSN 1468-0084

Forecasting Euro-Area Macroeconomic Variables Using a Factor Model Approach for Backdating

×
We suggest to use a factor model based backdating procedure to construct historical Euro-area macroeconomic time series data for the pre-Euro period. We<br />argue that this is a useful alternative to standard contemporaneous aggregation<br />methods. The paper investigates for a number of Euro-area variables whether<br />forecasts based on the factor-backdated data are more precise than those obtained<br />with standard area-wide data. A recursive pseudo-out-of-sample forecasting experiment using quarterly data is conducted. Our results suggest that some key variables (e.g. real GDP, inflation and long-term interest rate) can indeed be forecasted more precisely with the factor-backdated data.

Origin (projects)

  Zeng, Jing(2015): Forecasting Euro Area Macroeconomic Aggregate Variables

Forecasting Euro Area Macroeconomic Aggregate Variables

×
Zeng, Jing

Origin (projects)

  Zeng, Jing(2015): Combining Country-Specific Forecasts when Forecasting Euro Area Macroeconomic Aggregates

Combining Country-Specific Forecasts when Forecasting Euro Area Macroeconomic Aggregates

×
European Monetary Union (EMU) member countries' forecasts are often combined to obtain the forecasts of the Euro area macroeconomic aggregate variables. The aggregation weights which are used to produce the aggregates are often considered as combination weights. This paper investigates whether using different combination weights instead of the usual aggregation weights can help to provide more accurate forecasts. In this context, we examine the performance of equal weights, the least squares estimators of the weights, the combination method recently proposed by Hyndman et al. (2011) and the weights suggested by shrinkage methods. We find that some variables like real GDP and GDP deflator can be forecasted more precisely by using flexible combination weights. Furthermore, combining only forecasts of the three largest European countries helps to improve the forecasting performance. The persistence of the individual data seems to play an important role for the relative performance of the combination.

Origin (projects)

    Brüggemann, Ralf; Jentsch, Carsten; Trenkler, Carsten(2014): Inference in VARs with Conditional Heteroskedasticity of Unknown Form

Inference in VARs with Conditional Heteroskedasticity of Unknown Form

×
We derive a framework for asymptotically valid inference in stable vector autoregressive (VAR) models with conditional heteroskedasticity of unknown form. We prove a joint central limit theorem for the VAR slope parameter and innovation covariance parameter estimators and address bootstrap inference as well. Our results are important for correct inference on VAR statistics that depend both on the VAR slope and the variance parameters as e.g. in structural impulse response functions (IRFs). We also show that wild and pairwise bootstrap schemes fail in the presence of conditional heteroskedasticity if inference on (functions) of the unconditional variance parameters is of interest because they do not correctly replicate the relevant fourth moments' structure of the error terms. In contrast, the residual-based moving block bootstrap results in asymptotically valid inference. We illustrate the practical implications of our theoretical results by providing simulation evidence on the finite sample properties of different inference methods for IRFs. Our results point out that estimation uncertainty may increase dramatically in the presence of conditional heteroskedasticity. Moreover, most inference methods are likely to understate the true estimation uncertainty substantially in finite samples.

Origin (projects)

    Brüggemann, Ralf; Lütkepohl, Helmut(2013): Forecasting contemporaneous aggregates with stochastic aggregation weights International Journal of Forecasting ; 29 (2013), 1. - S. 60-68. - ISSN 0169-2070

Forecasting contemporaneous aggregates with stochastic aggregation weights

×
Many contemporaneously aggregated variables have stochastic aggregation weights. We compare different forecasts for such variables, including univariate forecasts of the aggregate, a multivariate forecast of the aggregate that uses information from the disaggregated components, a forecast which aggregates a multivariate forecast of the disaggregate components and the aggregation weights, and a forecast which aggregates univariate forecasts of individual disaggregate components and the aggregation weights. In empirical illustrations based on aggregate GDP and money stock series, we find forecast mean squared error reductions when information in the stochastic aggregation weights is used.

Origin (projects)

  Brüggemann, Ralf; Zeng, Jing(2012): Forecasting Euro-Area Macroeconomic Variables Using a Factor Model Approach for Backdating

Forecasting Euro-Area Macroeconomic Variables Using a Factor Model Approach for Backdating

×
We suggest to use a factor model based backdating procedure to construct historical Euro-area macroeconomic time series data for the pre-Euro period. We argue that this is a useful alternative to standard contemporaneous aggregation methods. The paper investigates for a number of Euro-area variables whether forecasts based on the factor-backdated data are more precise than those obtained with standard area-wide data. A recursive pseudo-out-of-sample forecasting experiment using quarterly data is conducted. Our results suggest that some key variables (e.g. real GDP, inflation and long-term interest rate) can indeed be forecasted more precisely with the factor-backdated data.

Origin (projects)

  Balabanova, Zlatina; Brüggemann, Ralf(2012): External Information and Monetary Policy Transmission in New EU Member States : Results from FAVAR Models

External Information and Monetary Policy Transmission in New EU Member States : Results from FAVAR Models

×
We investigate the e_ects of monetary policy shocks in the new European Union member states Czech Republic, Hungary, Poland and Slovakia. In contrast to existing studies, we explicitly account for external developments in European Monetary Union (EMU) countries and in other acceding countries. We do so by using factor-augmented vector-autoregressive models that employ the information from non-stationary factor time series. One set of VAR models includes factors obtained from a large cross-section of time series from EMU countries, while another set includes factors obtained from other acceding countries. We use cohesion analysis to facilitate the interpretation of the different factor time series. We find that including the EMU factors does not greatly affect the impulse response patterns in acceding countries. In contrast, including factors from other accession countries leads to substantial changes in impulse responses and to economically more plausible results. Overall, our analysis highlights that taking into account external economic developments properly is crucial for the analysis of monetary policy in the new EU member states.

Origin (projects)

Funding sources
NameProject no.DescriptionPeriod
Sachbeihilfe/Normalverfahren581/12DFGno information
Further information
Period: 20.06.2012 – 19.06.2014