Higher-order income dynamics with linked regression trees

Research output: Contribution to journal › Journal article › Research › peer-review

Standard

Higher-order income dynamics with linked regression trees. / Druedahl, Jeppe; Munk-nielsen, Anders.

In: The Econometrics Journal, Vol. 23, No. 3, 01.09.2020, p. S25-S58.

Research output: Contribution to journal › Journal article › Research › peer-review

Harvard

Druedahl, J & Munk-nielsen, A 2020, 'Higher-order income dynamics with linked regression trees', The Econometrics Journal, vol. 23, no. 3, pp. S25-S58. https://doi.org/10.1093/ectj/utaa026

APA

Druedahl, J., & Munk-nielsen, A. (2020). Higher-order income dynamics with linked regression trees. The Econometrics Journal, 23(3), S25-S58. https://doi.org/10.1093/ectj/utaa026

Vancouver

Druedahl J, Munk-nielsen A. Higher-order income dynamics with linked regression trees. The Econometrics Journal. 2020 Sep 1;23(3):S25-S58. https://doi.org/10.1093/ectj/utaa026

Author

Druedahl, Jeppe ; Munk-nielsen, Anders. / Higher-order income dynamics with linked regression trees. In: The Econometrics Journal. 2020 ; Vol. 23, No. 3. pp. S25-S58.

Bibtex

@article{bbbedbe5c34b4961929f7ba96484cab5,
title = "Higher-order income dynamics with linked regression trees",
abstract = "We propose a novel method for modelling income processes using machine learning. Our method links age-specific regression trees, and returns a discrete state process, which can easily be included in consumption-saving models without further discretizations. A central advantage of our approach is that it does not rely on any parametric assumptions, and because we build on existing machine learning tools it is furthermore easy to apply in practice. Using a 30-year panel of Danish males, we document rich higher-order income dynamics, including substantial skewness and high kurtosis of income levels and growth rates. We also find important changes in income risk over the life-cycle and the income distribution. Our estimated process matches these dynamics closely. Using a consumption-saving model, the implied welfare cost of income risk is more than 10% of income.",
keywords = "Faculty of Social Sciences, Income dynamics, higher-order income risk, consumption-saving, welfare cost of income risk, machine learning",
author = "Jeppe Druedahl and Anders Munk-nielsen",
year = "2020",
month = sep,
day = "1",
doi = "10.1093/ectj/utaa026",
language = "English",
volume = "23",
pages = "S25--S58",
journal = "Econometrics Journal",
issn = "1368-4221",
publisher = "Wiley",
number = "3",

}

RIS

TY - JOUR

T1 - Higher-order income dynamics with linked regression trees

AU - Druedahl, Jeppe

AU - Munk-nielsen, Anders

PY - 2020/9/1

Y1 - 2020/9/1

N2 - We propose a novel method for modelling income processes using machine learning. Our method links age-specific regression trees, and returns a discrete state process, which can easily be included in consumption-saving models without further discretizations. A central advantage of our approach is that it does not rely on any parametric assumptions, and because we build on existing machine learning tools it is furthermore easy to apply in practice. Using a 30-year panel of Danish males, we document rich higher-order income dynamics, including substantial skewness and high kurtosis of income levels and growth rates. We also find important changes in income risk over the life-cycle and the income distribution. Our estimated process matches these dynamics closely. Using a consumption-saving model, the implied welfare cost of income risk is more than 10% of income.

AB - We propose a novel method for modelling income processes using machine learning. Our method links age-specific regression trees, and returns a discrete state process, which can easily be included in consumption-saving models without further discretizations. A central advantage of our approach is that it does not rely on any parametric assumptions, and because we build on existing machine learning tools it is furthermore easy to apply in practice. Using a 30-year panel of Danish males, we document rich higher-order income dynamics, including substantial skewness and high kurtosis of income levels and growth rates. We also find important changes in income risk over the life-cycle and the income distribution. Our estimated process matches these dynamics closely. Using a consumption-saving model, the implied welfare cost of income risk is more than 10% of income.

KW - Faculty of Social Sciences

KW - Income dynamics

KW - higher-order income risk

KW - consumption-saving

KW - welfare cost of income risk

KW - machine learning

U2 - 10.1093/ectj/utaa026

DO - 10.1093/ectj/utaa026

M3 - Journal article

VL - 23

SP - S25-S58

JO - Econometrics Journal

JF - Econometrics Journal

SN - 1368-4221

IS - 3

ER -

ID: 252302032