Publications - Anders Bredahl Kock https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Bcontroller%5D=Publications&cHash=8e5a55fbb159195669431106134cfb53 en-us PURE Extension typo3support@science.au.dk (Web Department) 30 <![CDATA[Treatment recommendation with distributional targets]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=14709f32-f899-4f28-9790-d614b4bb404e&tx_pure_pure5%5BshowType%5D=pub&cHash=d45bd92c71bdc0b482d3ac1f6d2704d9 Kock, A. B., Preinerstorfer, D., Veliyev, B. Forskning Fri, 01 May 2020 19:19:20 +0200 14709f32-f899-4f28-9790-d614b4bb404e <![CDATA[Penalized Time Series Regression]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=bedf5f36-eff5-4f8c-87bf-959f89de5a52&tx_pure_pure5%5BshowType%5D=pub&cHash=0c1bfe1aabc659cbd1fac4b4898c2f92 Kock, A. B., Medeiros, M., Vasconcelos, G. This chapter covers penalized regression in the framework of linear time series models and reviews the most commonly used penalized estimators in applied work, namely Ridge Regression, the Least Absolute Shrinkage and Selection Operator (Lasso), the Elastic Net, the adaptive versions of the Lasso as well as Elastic Net and the group Lasso. Other penalties are briefly presented. We discuss theoretical properties such as consistent variable selection, the oracle property, and oracle inequalities and list time series models in which penalized estimators have been shown to possess these. Potentially problematic aspects of (some of) these properties are also discussed. Practical issues, such as the selection of the penalty parameters and available computer implementations, are also covered. A Monte Carlo simulation is presented in order to compare different penalties in terms of estimation precision, model selection capability, and forecasting performance. Finally, an application to forecasting US monthly inflation is presented.

]]>
Forskning Wed, 01 Jan 2020 19:19:20 +0100 bedf5f36-eff5-4f8c-87bf-959f89de5a52
<![CDATA[Functional Sequential Treatment Allocation with Covariates]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=aa1e2aa6-2734-4fa5-b976-ea4ec303365c&tx_pure_pure5%5BshowType%5D=pub&cHash=16b584e60aa93ef5a626d61204744ec8 Kock, A. B., Preinerstorfer, D., Veliyev, B. Forskning Wed, 01 Jan 2020 19:19:20 +0100 aa1e2aa6-2734-4fa5-b976-ea4ec303365c <![CDATA[Inference in partially identified models with many moment inequalities using Lasso]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=573e9185-4943-4ae0-9d1f-607570a48e55&tx_pure_pure5%5BshowType%5D=pub&cHash=5eddfa54c3500e016b70dfcf09f59a54 Bugni, F. A., Caner, M., Bredahl Kock, A., Lahiri, S. This paper considers inference in a partially identified moment (in)equality model with many moment inequalities. We propose a novel two-step inference procedure that combines the methods proposed by Chernozhukov et al. (2018a) (Chernozhukov et al., 2018a, hereafter) with a first step moment inequality selection based on the Lasso. Our method controls asymptotic size uniformly, both in the underlying parameter and the data distribution. Also, the power of our method compares favorably with that of the corresponding two-step method in Chernozhukov et al. (2018a) for large parts of the parameter space, both in theory and in simulations. Finally, we show that our Lasso-based first step can be implemented by thresholding standardized sample averages, and so it is straightforward to implement.

]]>
Forskning Fri, 01 May 2020 19:19:21 +0200 573e9185-4943-4ae0-9d1f-607570a48e55
<![CDATA[Power in High-Dimensional Testing Problems]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=9cbf51a1-cd0e-4bb9-9389-ad1647550d9e&tx_pure_pure5%5BshowType%5D=pub&cHash=f4b9e7f085e019181d32c6087274e1b4 Kock, A. B., Preinerstorfer, D. Fan, Liao, and Yao (2015) recently introduced a remarkable method for increasing the asymptotic power of tests in high-dimensional testing problems. If applicable to a given test, their power enhancement principle leads to an improved test that has the same asymptotic size, has uniformly non-inferior asymptotic power, and is consistent against a strictly broader range of alternatives than the initially given test. We study under which conditions this method can be applied and show the following: In asymptotic regimes where the dimensionality of the parameter space is fixed as sample size increases, there often exist tests that cannot be further improved with the power enhancement principle. However, when the dimensionality of the parameter space increases sufficiently slowly with sample size and a marginal local asymptotic normality (LAN) condition is satisfied, every test with asymptotic size smaller than 1 can be improved with the power enhancement principle. While the marginal LAN condition alone does not allow one to extend the latter statement to all rates at which the dimensionality increases with sample size, we give sufficient conditions under which this is the case.

]]>
Forskning Wed, 01 May 2019 19:19:21 +0200 9cbf51a1-cd0e-4bb9-9389-ad1647550d9e
<![CDATA[Functional Sequential Treatment Allocation]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=a1dc3441-a12c-4ec0-aa24-9b95bd2e113f&tx_pure_pure5%5BshowType%5D=pub&cHash=c8357df202a9a81a580ad1ce9285610b Kock, A. B., Preinerstorfer, D., Veliyev, B. Forskning Wed, 01 Jan 2020 19:19:21 +0100 a1dc3441-a12c-4ec0-aa24-9b95bd2e113f <![CDATA[Uniform inference in highdimensional dynamic panel data models with approximately sparse fixed effects]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=be3ac9c1-dbcd-45d0-b653-e2ee650f0b28&tx_pure_pure5%5BshowType%5D=pub&cHash=1131edf71ef578454730c8cd9e5ed504 Kock, A. B., Tang, H. We establish oracle inequalities for a version of the Lasso in high-dimensional fixed effects dynamic panel data models. The inequalities are valid for the coefficients of the dynamic and exogenous regressors. Separate oracle inequalities are derived for the fixed effects. Next, we show how one can conduct uniformly valid inference on the parameters of the model and construct a uniformly valid estimator of the asymptotic covariance matrix which is robust to conditional heteroskedasticity in the error terms. Allowing for conditional heteroskedasticity is important in dynamic models as the conditional error variance may be nonconstant over time and depend on the covariates. Furthermore, our procedure allows for inference on high-dimensional subsets of the parameter vector of an increasing cardinality. We show that the confidence bands resulting from our procedure are asymptotically honest and contract at the optimal rate. This rate is different for the fixed effects than for the remaining parts of the parameter vector.

]]>
Forskning Mon, 01 Apr 2019 19:19:21 +0200 be3ac9c1-dbcd-45d0-b653-e2ee650f0b28
<![CDATA[Asymptotically honest confidence regions for high dimensional parameters by the desparsified conservative Lasso]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=104e03c0-42f0-43c4-b154-b6f71c6dbbae&tx_pure_pure5%5BshowType%5D=pub&cHash=db0810d6c57544f7e0ebf6c8e74f142f Caner, M., Kock, A. B. In this paper we consider the conservative Lasso which we argue penalizes more correctly than the Lasso and show how it may be desparsified in the sense of van de Geer et al. (2014) in order to construct asymptotically honest (uniform) confidence bands. In particular, we develop an oracle inequality for the conservative Lasso only assuming the existence of a certain number of moments. This is done by means of the Marcinkiewicz–Zygmund inequality. We allow for heteroskedastic non-subgaussian error terms and covariates. Next, we desparsify the conservative Lasso estimator and derive the asymptotic distribution of tests involving an increasing number of parameters. Our simulations reveal that the desparsified conservative Lasso estimates the parameters more precisely than the desparsified Lasso, has better size properties and produces confidence bands with superior coverage rates.

]]>
Forskning Mon, 01 Jan 2018 19:19:21 +0100 104e03c0-42f0-43c4-b154-b6f71c6dbbae
<![CDATA[Oracle inequalities, variable selection and uniform inference in high-dimensional correlated random effects panel data models]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=5e1ce05a-71ca-44c0-aa71-15bcab64ae60&tx_pure_pure5%5BshowType%5D=pub&cHash=77c4ded87e0a58950ad6d38012753aab Kock, A. B. In this paper we study high-dimensional correlated random effects panel data models. Our setting is useful as it allows including time invariant covariates as under random effects yet allows for correlation between covariates and unobserved heterogeneity as under fixed effects. We use the Mundlak-Chamberlain device to model this correlation. Allowing for a flexible correlation structure naturally leads to a high dimensional model in which least squares estimation easily becomes infeasible with even a moderate number of explanatory variables.

Imposing a combination of sparsity and weak sparsity on the parameters of the model we first establish an oracle inequality for the Lasso. This is valid even when the error terms are-heteroskedastic and no structure is imposed on the time series dependence of the error terms.

Next, we provide upper bounds on the sup-norm estimation error of the Lasso. As opposed to the classical l(1)- and l(2)-bounds the sup-norm bounds do not directly depend on the unknown degree of sparsity and are thus well suited for thresholding the Lasso for variable selection. We provide sufficient conditions under which thresholding results in consistent model selection. Pointwise valid asymptotic inference is established for a post-thresholding estimator. Finally, we show how the Lasso can be desparsified in the correlated random effects setting and how this leads to uniformly valid inference even in the presence of heteroskedasticity and autocorrelated error terms. (C) 2016 Elsevier B.V. All rights reserved.

]]>
Forskning Tue, 01 Nov 2016 19:19:21 +0100 5e1ce05a-71ca-44c0-aa71-15bcab64ae60
<![CDATA[Sharp threshold detection based on sup-norm error rates in high-dimensional models]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=1f2a22ad-d939-4d9b-93c1-820b3026282e&tx_pure_pure5%5BshowType%5D=pub&cHash=fe6d961d229ac49f6d2cbb71934a2399 Callot, L., Caner, M., Kock, A. B., Riquelme, J. A. We propose a new estimator, the thresholded scaled Lasso, in high-dimensional threshold regressions. First, we establish an upper bound on the ℓ estimation error of the scaled Lasso estimator of Lee, Seo, and Shin. This is a nontrivial task as the literature on high-dimensional models has focused almost exclusively on ℓ 1 and ℓ 2 estimation errors. We show that this sup-norm bound can be used to distinguish between zero and nonzero coefficients at a much finer scale than would have been possible using classical oracle inequalities. Thus, our sup-norm bound is tailored to consistent variable selection via thresholding. Our simulations show that thresholding the scaled Lasso yields substantial improvements in terms of variable selection. Finally, we use our estimator to shed further empirical light on the long-running debate on the relationship between the level of debt (public and private) and GDP growth. Supplementary materials for this article are available online.

]]>
Forskning Sun, 01 Jan 2017 19:19:21 +0100 1f2a22ad-d939-4d9b-93c1-820b3026282e
<![CDATA[Inference in partially identified models with many moment inequalities using Lasso]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=27f7a83f-f0ee-4604-8e47-2eabec87c7f3&tx_pure_pure5%5BshowType%5D=pub&cHash=ae7bb45723ac0c440214a4ef03616f69 Bugni, F. A., Caner, M., Kock, A. B., Lahiri, S. Forskning Wed, 27 Apr 2016 19:19:21 +0200 27f7a83f-f0ee-4604-8e47-2eabec87c7f3 <![CDATA[Modeling and Forecasting Large Realized Covariance Matrices and Portfolio Choice]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=7d357cc0-92b7-4d83-b9af-8617d974cf33&tx_pure_pure5%5BshowType%5D=pub&cHash=6a4e82dfac51f0cdc47e85cb358b74ae Callot, L., Kock, A. B., Medeiros, M. Forskning Sun, 01 Jan 2017 19:19:21 +0100 7d357cc0-92b7-4d83-b9af-8617d974cf33 <![CDATA[Consistent and Conservative Model Selection with the Adaptive LASSO in Stationary and Nonstationary Autoregressions]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=407f5a4f-dedf-47a4-9ad6-47c7f9f42559&tx_pure_pure5%5BshowType%5D=pub&cHash=cdcd634ae8783190bace98cfdc69253d Kock, A. B. However, it is also shown that the adaptive Lasso has no power against shrinking alternatives of the form c/T if it is tuned to perform consistent model selection. We show that if the adaptive Lasso is tuned to perform conservative model selection it has power even against shrinking alternatives of this form and compare it to the plain Lasso.]]> Forskning Fri, 01 Jan 2016 19:19:21 +0100 407f5a4f-dedf-47a4-9ad6-47c7f9f42559 <![CDATA[Oracle Inequalities for Convex Loss Functions with Non-Linear Targets]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=b077b468-1310-40c6-861e-210d3e74a3b8&tx_pure_pure5%5BshowType%5D=pub&cHash=dd832f200bce62c0a819c40d59c959e2 Caner, M., Kock, A. B. Forskning Fri, 01 Jan 2016 19:19:21 +0100 b077b468-1310-40c6-861e-210d3e74a3b8 <![CDATA[Sharp Threshold Detection Based on Sup-norm Error rates in High-dimensional Models]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=a21f930c-af5b-4be1-8771-3b2926896a16&tx_pure_pure5%5BshowType%5D=pub&cHash=1a248a6c5f91bef5d40d8db619db240a Callot, L., Caner, M., Kock, A. B., Riquelme, J. A. Forskning Tue, 17 Feb 2015 19:19:21 +0100 a21f930c-af5b-4be1-8771-3b2926896a16 <![CDATA[Oracle Efficient Estimation and Forecasting With the Adaptive Lasso and the Adaptive Group Lasso in Vector Autoregressions]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=a6053d74-bbb1-474f-9385-93a7794f33de&tx_pure_pure5%5BshowType%5D=pub&cHash=6a6bc48b5edee342e2ccea36d2fdd603 Kock, A. B., Callot, L. Forskning Wed, 01 Jan 2014 19:19:21 +0100 a6053d74-bbb1-474f-9385-93a7794f33de <![CDATA[Lassoing the Determinants of Retirement]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=b16f62f3-588e-4c65-a8fd-58bf31465377&tx_pure_pure5%5BshowType%5D=pub&cHash=1b7e3338ebc7805310e87892a48f9961 Kallestrup-Lamb, M., Kock, A. B., Kristensen, J. T. Forskning Fri, 01 Jan 2016 19:19:21 +0100 b16f62f3-588e-4c65-a8fd-58bf31465377 <![CDATA[Oracle Inequalities for High-Dimensional Vector Autoregressions]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=aa947c3e-1472-4039-9cc0-99a07c96bcd4&tx_pure_pure5%5BshowType%5D=pub&cHash=d323cc07e0c171b78a547084280ee1fc Kock, A. B., Callot, L. Forskning Mon, 01 Jun 2015 19:19:21 +0200 aa947c3e-1472-4039-9cc0-99a07c96bcd4 <![CDATA[Forecasting macroeconomic variables using neural network models and three automated model selection techniques]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=94404bd7-e522-4f44-b380-d06e6a5432da&tx_pure_pure5%5BshowType%5D=pub&cHash=6067ea2b2fb7f4a07dc8f274c82a9b3e Kock, A. B., Teräsvirta, T. Forskning Fri, 01 Jan 2016 19:19:21 +0100 94404bd7-e522-4f44-b380-d06e6a5432da <![CDATA[Inference in High-dimensional Dynamic Panel Data Models]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=baa6b304-1fd3-4c02-a3f8-6d8cc1f2e0f5&tx_pure_pure5%5BshowType%5D=pub&cHash=373d24d73065ad4bce79d045f321de1c Kock, A. B., Tang, H. Forskning Wed, 01 Jan 2014 19:19:21 +0100 baa6b304-1fd3-4c02-a3f8-6d8cc1f2e0f5 <![CDATA[Estimation and Forecasting of Large Realized Covariance Matrices and Portfolio Choice]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=2d29d526-5344-4aea-86d1-eca138dc15ac&tx_pure_pure5%5BshowType%5D=pub&cHash=552d0a20d06a7fa86472e26b381c8bb5 Callot, L., Kock, A. B., Medeiros, M. C. Forskning Mon, 17 Nov 2014 19:19:21 +0100 2d29d526-5344-4aea-86d1-eca138dc15ac <![CDATA[Asymptotically Honest Confidence Regions for High Dimensional]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=b7b8b2a7-2114-410a-8f9e-b71b48dbcd4d&tx_pure_pure5%5BshowType%5D=pub&cHash=bbd4c13e9f0420e5d8abee67e241d85b Caner, M., Kock, A. B. Forskning Tue, 21 Oct 2014 19:19:21 +0200 b7b8b2a7-2114-410a-8f9e-b71b48dbcd4d <![CDATA[Oracle Inequalities for Convex Loss Functions with Non-Linear Targets]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=0f4afe3f-ec85-4b3a-95de-78e2c153e235&tx_pure_pure5%5BshowType%5D=pub&cHash=5f7092a9316753b5b4576d638afc8c5e Caner, M., Kock, A. B. Forskning Fri, 20 Dec 2013 19:19:21 +0100 0f4afe3f-ec85-4b3a-95de-78e2c153e235 <![CDATA[Forecasting the Finnish Consumer Price Inflation Using Artificial Neural Network Models and Three Automated Model Selection Techniques]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=0395ceb0-c523-4b05-9aa6-ef4f5e924aa1&tx_pure_pure5%5BshowType%5D=pub&cHash=542197a387de47761c1584a488bc5e0c Kock, A. B., Teräsvirta, T. Forskning Tue, 01 Jan 2013 19:19:21 +0100 0395ceb0-c523-4b05-9aa6-ef4f5e924aa1 <![CDATA[Lassoing the Determinants of Retirement]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=0803377f-93b0-441b-9eb4-4afc049d9dd8&tx_pure_pure5%5BshowType%5D=pub&cHash=51accc87f748ece21ae64c083078a381 Kallestrup-Lamb, M., Kock, A. B., Kristensen, J. T. Forskning Mon, 01 Jul 2013 19:19:21 +0200 0803377f-93b0-441b-9eb4-4afc049d9dd8 <![CDATA[Oracle inequalities for high-dimensional panel data models]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=ad96c08e-708a-47c5-8531-3c22945b26a1&tx_pure_pure5%5BshowType%5D=pub&cHash=62ecc437d5acd6ab61fe5d012ba9fddd Kock, A. B. Forskning Thu, 13 Jun 2013 19:19:21 +0200 ad96c08e-708a-47c5-8531-3c22945b26a1 <![CDATA[Forecasting performances of three automated modelling techniques during the economic crisis 2007-2009]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=13400638-9950-4e3d-bba2-4272f87ec094&tx_pure_pure5%5BshowType%5D=pub&cHash=eec5fc410603a129ca9ae71ea548839e Kock, A. B., Teräsvirta, T. Forskning Wed, 01 Jan 2014 19:19:21 +0100 13400638-9950-4e3d-bba2-4272f87ec094 <![CDATA[Oracle Efficient Estimation and Forecasting with the Adaptive LASSO and the Adaptive Group LASSO in Vector Autoregressions]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=77f3e3cc-9c16-4240-a61f-84cf4300f051&tx_pure_pure5%5BshowType%5D=pub&cHash=a02baeeebc9e8197818b855ba6dd6a9f Kock, A. B., Callot, L. We evaluate the forecasting accuracy of these estimators for a large set of macroeconomic variables. The Lasso is found to be the most precise procedure overall. The adaptive and the adaptive group Lasso are less stable but mostly perform at par with the common factor models.]]> Forskning Thu, 06 Sep 2012 19:19:21 +0200 77f3e3cc-9c16-4240-a61f-84cf4300f051 <![CDATA[Oracle Inequalities for High Dimensional Vector Autoregressions]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=9ace80f8-e699-4bb8-a94b-a48e1e177de8&tx_pure_pure5%5BshowType%5D=pub&cHash=6000854c67d68cbcf06bc78d90e46859 Callot, L., Kock, A. B. Forskning Mon, 07 May 2012 19:19:21 +0200 9ace80f8-e699-4bb8-a94b-a48e1e177de8 <![CDATA[Oracle Efficient Variable Selection in Random and Fixed Effects Panel Data Models]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=61cc90bf-b09c-4d73-a02a-05650e6655d1&tx_pure_pure5%5BshowType%5D=pub&cHash=b362539e9f9c5a833898b6b9ee87c8fd Kock, A. B. Forskning Tue, 01 Jan 2013 19:19:21 +0100 61cc90bf-b09c-4d73-a02a-05650e6655d1 <![CDATA[Forecasting with Universal Approximators and a Learning Algorithm]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=fc5593f4-cd68-43be-8276-a78faabd5503&tx_pure_pure5%5BshowType%5D=pub&cHash=c4b422ba33269221a75d912ef49cb7cc Kock, A. B. performed. The practical performance will be investigated by considering various monthly postwar macroeconomic data sets for the G7 as well as the Scandinavian countries.]]> Forskning Sat, 01 Jan 2011 19:19:21 +0100 fc5593f4-cd68-43be-8276-a78faabd5503 <![CDATA[On the Oracle Property of the Adaptive LASSO in Stationary and Nonstationary Autoregressions]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=e51c49f8-41cd-43a7-a90a-8dd80d8a5d98&tx_pure_pure5%5BshowType%5D=pub&cHash=b55d086b47587230f23b2b561d3b2fd1 Kock, A. B. Forskning Fri, 03 Feb 2012 19:19:21 +0100 e51c49f8-41cd-43a7-a90a-8dd80d8a5d98 <![CDATA[Forecasting and Oracle Efficient Econometrics]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=68bf3c32-5a2e-4546-9650-6048ca585ba7&tx_pure_pure5%5BshowType%5D=pub&cHash=41c4e52870d7b173cf35bb0845230889 Kock, A. B. Forskning Sat, 01 Jan 2011 19:19:21 +0100 68bf3c32-5a2e-4546-9650-6048ca585ba7 <![CDATA[Forecasting performance of three automated modelling techniques during the economic crisis 2007-2009]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=831870e1-04a0-489a-8b5e-b9d85b995182&tx_pure_pure5%5BshowType%5D=pub&cHash=6f0bd911746c96ac1c73a4640097993c Kock, A. B., Teräsvirta, T. problem. To this end we employ three automatic modelling devices. One of them is White’s QuickNet, but we also consider Autometrics, well known to time series econometricians, and the Marginal Bridge Estimator, better known to statisticians and microeconometricians.The performance of these three model selectors is compared by looking at the accuracy of the forecasts of the estimated neural network models. We apply the neural network model and the three modelling
techniques to monthly industrial production and unemployment series of the G7 countries and the four Scandinavian ones, and focus on forecasting during the economic crisis 2007–2009. Forecast accuracy is measured by the root mean square forecast error. Hypothesis testing is also used to compare the performance of the different techniques with each other.]]>
Forskning Sat, 01 Jan 2011 19:19:21 +0100 831870e1-04a0-489a-8b5e-b9d85b995182
<![CDATA[Forecasting Macroeconomic Variables using Neural Network Models and Three Automated Model Selection Techniques]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=9af05103-c64f-41d8-984a-dae7732d1dfe&tx_pure_pure5%5BshowType%5D=pub&cHash=d281daa07c540a78536d690a330f34ba Kock, A. B., Teräsvirta, T. fact, their parameters are not even globally identified. Recently, White (2006) presented a solution that amounts to converting the specification and nonlinear estimation problem into a linear model selection and estimation problem. He called this procedure the QuickNet and we shall compare its performance to two other procedures which are built on the linearisation idea: the Marginal Bridge Estimator and Autometrics. Second, one must decide whether forecasting should be carried out recursively or directly. Comparisons of these two methodss exist for linear models and here these comparisons are extended to neural networks.
Finally, a nonlinear model such as the neural network model is not appropriate if the data is generated by a linear mechanism. Hence, it might be appropriate to test the null of linearity prior to building a nonlinear model. We investigate whether this kind of pretesting improves the forecast accuracy compared to the case where this is not done. ]]>
Forskning Sat, 01 Jan 2011 19:19:21 +0100 9af05103-c64f-41d8-984a-dae7732d1dfe
<![CDATA[Forecasting with nonlinear time series models]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=3eb0428d-a583-4af5-bcae-d2e9f0045a52&tx_pure_pure5%5BshowType%5D=pub&cHash=1dc783217616103b34bb0162a221516a Kock, A. B., Teräsvirta, T. Forskning Wed, 01 Jun 2011 19:19:21 +0200 3eb0428d-a583-4af5-bcae-d2e9f0045a52 <![CDATA[Oracle Efficient Variable Selection in Random and Fixed Effects Panel Data Models]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=a04a1ea0-b682-11df-b4b4-000ea68e967b&tx_pure_pure5%5BshowType%5D=pub&cHash=5623d37ee6f13c879c3c0bec2fb87684 Kock, A. B. estimator is oracle efficient. It can correctly distinguish between relevant and irrelevant variables and the asymptotic distribution of the estimators of the coefficients of the relevant variables is the same as if only these had been included in the model, i.e. as if an oracle had revealed the true model prior to estimation. In the case of more explanatory variables than observations, we prove that the Marginal Bridge estimator can asymptotically correctly distinguish between relevant and irrelevant explanatory variables. We do this without restricting the dependence between covariates and without assuming sub Gaussianity of the error terms thereby generalizing the results of Huang et al. (2008). Furthermore, the number of relevant variables is allowed to be larger than the sample size.
]]>
Forskning Fri, 01 Jan 2010 19:19:21 +0100 a04a1ea0-b682-11df-b4b4-000ea68e967b
<![CDATA[Forecasting with nonlinear time series models]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=1d32b830-f91b-11de-9c17-000ea68e967b&tx_pure_pure5%5BshowType%5D=pub&cHash=836db7398d11929fbca547142800be3b Kock, A. B., Teräsvirta, T. parametric models. Several such models popular in time series econo-
metrics are presented and some of their properties discussed. This in-
cludes two models based on universal approximators: the Kolmogorov-
Gabor polynomial model and two versions of a simple artificial neural
network model. Techniques for generating multi-period forecasts from
nonlinear models recursively are considered, and the direct (non-recursive)
method for this purpose is mentioned as well. Forecasting with com-
plex dynamic systems, albeit less frequently applied to economic fore-
casting problems, is briefly highlighted. A number of large published
studies comparing macroeconomic forecasts obtained using different
time series models are discussed, and the paper also contains a small
simulation study comparing recursive and direct forecasts in a partic-
ular case where the data-generating process is a simple artificial neural
network model. Suggestions for further reading conclude the paper.]]>
Forskning Fri, 01 Jan 2010 19:19:21 +0100 1d32b830-f91b-11de-9c17-000ea68e967b
<![CDATA[Forecasting with Universal Approximators and a Learning Algorithm]]> https://econ.au.dk/da/research/researchcentres/creates/people/research-fellows/anders-bredahl-kock?tx_pure_pure5%5Baction%5D=single&tx_pure_pure5%5Bcontroller%5D=Publications&tx_pure_pure5%5Bid%5D=7f2a59c0-3e19-11de-8dc9-000ea68e967b&tx_pure_pure5%5BshowType%5D=pub&cHash=810508fc0f24e5140517ecd7eebcf7a8 Kock, A. B. forecasting. They are the Artificial Neural Networks, the Kolmogorov-
Gabor polynomials, as well as the Elliptic Basis Function Networks.
Even though forecast combination has a long history in
econometrics focus has not been on proving loss bounds for the
combination rules applied. We apply the Weighted Average Algorithm
(WAA) of Kivinen and Warmuth (1999) for which such loss
bounds exist. Specifically, one can bound the worst case performance
of the WAA compared to the performance of the best single
model in the set of models combined from. The use of universal
approximators along with a combination scheme for which explicit
loss bounds exist should give a solid theoretical foundation to the
way the forecasts are performed. The practical performance will
be investigated by considering various monthly postwar macroeconomic
data sets for the G7 as well as the Scandinavian countries.]]>
Forskning Thu, 01 Jan 2009 19:19:21 +0100 7f2a59c0-3e19-11de-8dc9-000ea68e967b