Panel threshold regressions with latent group structures

Miao, Ke; Su, Liangjun; Wang, Wendun

doi:10.1016/j.jeconom.2019.07.006

Cited by 21 publications

(34 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Secondly, the obtained latent group structure is usually informative for empirical analysis, such as finding possible unobserved confounders and performing secondary analysis. Another benefit is that the latent group structure allows a convenient way to incorporate the spatial dependence information in the model and in turn helps improve the accuracy/efficiency of model fit and interpretation (Miao et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

“…This method was further extended to allow both group-varying intercept and slopes by Su et al (2016). Several more complicated models have also been proposed with different features, e.g., group-specific time patterns (Bonhomme and Manresa, 2015), time-varying grouped coefficients (Su et al, 2019), and group-varying threshold variables (Miao et al, 2020). Despite the success of those frequentist approaches, limited effort has been made under the Bayesian framework until very recently (Zhang, 2020;Ma et al, 2020;Hu et al, 2021;Geng and Hu, 2021).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Identifying Latent Groups in Spatial Panel Data Using a Markov Random Field Constrained Product Partition Model

Pan¹,

Hu²,

Shen³

2024

STAT SINICA

View full text Add to dashboard Cite

Understanding the heterogeneity over spatial locations is an important problem that has been widely studied in many applications such as economics and environmental science. In this paper, we focus on regression models for spatial panel data analysis, where repeated measurements are collected over time at various spatial locations. We propose a novel class of nonparametric priors that combines Markov random field (MRF) with the product partition model (PPM) and show that the resulting prior, called by MRF-PPM, is capable of identifying the latent group structures among the spatial locations while efficiently utilizing the spatial dependence information. We derive a closed-form conditional distribution for the proposed prior and introduce a new way to compute the marginal likelihood that renders efficient Bayesian inference. We further study the theoretical properties of the proposed MRF-PPM prior and show a clustering consistency result for the posterior distribution. We demonstrate the excellent empirical performance of our method via extensive simulation studies and applications to a US precipitation data and a California median household income data study.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Identifying Latent Groups in Spatial Panel Data Using a Markov Random Field Constrained Product Partition Model

Pan¹,

Hu²,

Shen³

2024

STAT SINICA

View full text Add to dashboard Cite

show abstract

“…Originally proposed for linear regression models, this approach has also been extended to quantile regression models by Zhang et al (2019a) and Leng et al (2021). Further advancement of this literature has considered time varying group membership, for example Miao et al (2020), Okui and Wang (2021) and Lumsdaine et al (2021).…”

Section: Introductionmentioning

confidence: 99%

Group structure estimation for panel data -- a general approach

Lei¹,

Gu²,

Volgushev³

2022

Preprint

View full text Add to dashboard Cite

Consider a panel data setting where repeated observations on individuals are available. Often it is reasonable to assume that there exist groups of individuals that share similar effects of observed characteristics, but the grouping is typically unknown in advance. We propose a novel approach to estimate such unobserved groupings for general panel data models. Our method explicitly accounts for the uncertainty in individual parameter estimates and remains computationally feasible with a large number of individuals and/or repeated measurements on each individual. The developed ideas can be applied even when individual-level data are not available and only parameter estimates together with some quantification of uncertainty are given to the researcher.

show abstract

“…Secondly, the obtained latent group structure is usually informative for empirical analysis, such as finding possible unobserved confounders and performing secondary analysis. Another benefit is that the latent group structure for spatial panel data in general tends to coincide with the geographic locations of the study (Miao et al, 2020), which automatically provides a useful way to incorporate the spatial dependence information and hence improve the efficiency/accuracy of the studied model.…”

Section: Introductionmentioning

confidence: 99%

“…This method was further extended to allow both group-varying intercept and slopes by Su et al (2016). Several more complicated models have also been proposed with different features, e.g., group-specific time patterns (Bonhomme and Manresa, 2015), time-varying grouped coefficients (Su et al, 2019) and group-varying threshold variables (Miao et al, 2020). Despite the success of those frequentist approaches, limited effort has been made under the Bayesian framework until very recently (Zhang, 2020).…”

Section: Introductionmentioning

confidence: 99%

Identifying latent groups in spatial panel data using a Markov random field constrained product partition model

Pan¹,

Hu²,

Shen³

2020

Preprint

View full text Add to dashboard Cite

Understanding the heterogeneity over spatial locations is an important problem that has been widely studied in many applications such as economics and environmental science. In this paper, we focus on regression models for spatial panel data analysis, where repeated measurements are collected over time at various spatial locations. We propose a novel class of nonparametric priors that combines Markov random field (MRF) with the product partition model (PPM), and show that the resulting prior, called by MRF-PPM, is capable of identifying the latent group structure among the spatial locations while efficiently utilizing the spatial dependence information. We derive a closed-form conditional distribution for the proposed prior and introduce a new way to compute the marginal likelihood that renders efficient Bayesian inference. We further study the theoretical properties of the proposed MRF-PPM prior and show a clustering consistency result for the posterior distribution. We demonstrate the excellent empirical performance of our method via extensive simulation studies and applications to a US precipitation data and a California median household income data study.

show abstract

Panel threshold regressions with latent group structures

Cited by 21 publications

References 42 publications

Identifying Latent Groups in Spatial Panel Data Using a Markov Random Field Constrained Product Partition Model

Identifying Latent Groups in Spatial Panel Data Using a Markov Random Field Constrained Product Partition Model

Group structure estimation for panel data -- a general approach

Identifying latent groups in spatial panel data using a Markov random field constrained product partition model

Contact Info

Product

Resources

About