CREATES

Diffusion Indexes With Sparse Loadings

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articlepeer-review

Standard

Diffusion Indexes With Sparse Loadings. / Kristensen, Johannes Tang.

In: Journal of Business and Economic Statistics, Vol. 35, No. 3, 2017, p. 434-451.

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articlepeer-review

Harvard

Kristensen, JT 2017, 'Diffusion Indexes With Sparse Loadings', Journal of Business and Economic Statistics, vol. 35, no. 3, pp. 434-451. https://doi.org/10.1080/07350015.2015.1084308

APA

Kristensen, J. T. (2017). Diffusion Indexes With Sparse Loadings. Journal of Business and Economic Statistics, 35(3), 434-451. https://doi.org/10.1080/07350015.2015.1084308

CBE

Kristensen JT. 2017. Diffusion Indexes With Sparse Loadings. Journal of Business and Economic Statistics. 35(3):434-451. https://doi.org/10.1080/07350015.2015.1084308

MLA

Vancouver

Kristensen JT. Diffusion Indexes With Sparse Loadings. Journal of Business and Economic Statistics. 2017;35(3):434-451. https://doi.org/10.1080/07350015.2015.1084308

Author

Kristensen, Johannes Tang. / Diffusion Indexes With Sparse Loadings. In: Journal of Business and Economic Statistics. 2017 ; Vol. 35, No. 3. pp. 434-451.

Bibtex

@article{d16a6dd0f6724c4abf6fdb4f987cc642,
title = "Diffusion Indexes With Sparse Loadings",
abstract = "The use of large-dimensional factor models in forecasting has received much attention in the literature with the consensus being that improvements on forecasts can be achieved when comparing with standard models. However, recent contributions in the literature have demonstrated that care needs to be taken when choosing which variables to include in the model. A number of different approaches to determining these variables have been put forward. These are, however, often based on ad hoc procedures or abandon the underlying theoretical factor model. In this article, we will take a different approach to the problem by using the least absolute shrinkage and selection operator (LASSO) as a variable selection method to choose between the possible variables and thus obtain sparse loadings from which factors or diffusion indexes can be formed. This allows us to build a more parsimonious factor model that is better suited for forecasting compared to the traditional principal components (PC) approach. We provide an asymptotic analysis of the estimator and illustrate its merits empirically in a forecasting experiment based on U.S. macroeconomic data. Overall we find that compared to PC we obtain improvements in forecasting accuracy and thus find it to be an important alternative to PC. Supplementary materials for this article are available online.",
keywords = "Factor models, Forecasting, LASSO, Principal components analysis, PRINCIPAL COMPONENT ANALYSIS, APPROXIMATE FACTOR MODELS, SELECTION, REGULARIZATION, ESTIMATORS, REGRESSION, FORECASTS, NUMBER",
author = "Kristensen, {Johannes Tang}",
year = "2017",
doi = "10.1080/07350015.2015.1084308",
language = "English",
volume = "35",
pages = "434--451",
journal = "Journal of Business and Economic Statistics",
issn = "0735-0015",
publisher = "Taylor & Francis Inc.",
number = "3",

}

RIS

TY - JOUR

T1 - Diffusion Indexes With Sparse Loadings

AU - Kristensen, Johannes Tang

PY - 2017

Y1 - 2017

N2 - The use of large-dimensional factor models in forecasting has received much attention in the literature with the consensus being that improvements on forecasts can be achieved when comparing with standard models. However, recent contributions in the literature have demonstrated that care needs to be taken when choosing which variables to include in the model. A number of different approaches to determining these variables have been put forward. These are, however, often based on ad hoc procedures or abandon the underlying theoretical factor model. In this article, we will take a different approach to the problem by using the least absolute shrinkage and selection operator (LASSO) as a variable selection method to choose between the possible variables and thus obtain sparse loadings from which factors or diffusion indexes can be formed. This allows us to build a more parsimonious factor model that is better suited for forecasting compared to the traditional principal components (PC) approach. We provide an asymptotic analysis of the estimator and illustrate its merits empirically in a forecasting experiment based on U.S. macroeconomic data. Overall we find that compared to PC we obtain improvements in forecasting accuracy and thus find it to be an important alternative to PC. Supplementary materials for this article are available online.

AB - The use of large-dimensional factor models in forecasting has received much attention in the literature with the consensus being that improvements on forecasts can be achieved when comparing with standard models. However, recent contributions in the literature have demonstrated that care needs to be taken when choosing which variables to include in the model. A number of different approaches to determining these variables have been put forward. These are, however, often based on ad hoc procedures or abandon the underlying theoretical factor model. In this article, we will take a different approach to the problem by using the least absolute shrinkage and selection operator (LASSO) as a variable selection method to choose between the possible variables and thus obtain sparse loadings from which factors or diffusion indexes can be formed. This allows us to build a more parsimonious factor model that is better suited for forecasting compared to the traditional principal components (PC) approach. We provide an asymptotic analysis of the estimator and illustrate its merits empirically in a forecasting experiment based on U.S. macroeconomic data. Overall we find that compared to PC we obtain improvements in forecasting accuracy and thus find it to be an important alternative to PC. Supplementary materials for this article are available online.

KW - Factor models

KW - Forecasting

KW - LASSO

KW - Principal components analysis

KW - PRINCIPAL COMPONENT ANALYSIS

KW - APPROXIMATE FACTOR MODELS

KW - SELECTION

KW - REGULARIZATION

KW - ESTIMATORS

KW - REGRESSION

KW - FORECASTS

KW - NUMBER

U2 - 10.1080/07350015.2015.1084308

DO - 10.1080/07350015.2015.1084308

M3 - Journal article

VL - 35

SP - 434

EP - 451

JO - Journal of Business and Economic Statistics

JF - Journal of Business and Economic Statistics

SN - 0735-0015

IS - 3

ER -