Modeling diurnal hormone profiles by hierarchical state space models

Ziyue Liu, Wensheng Guo

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Adrenocorticotropic hormone (ACTH) diurnal patterns contain both smooth circadian rhythms and pulsatile activities. How to evaluate and compare them between different groups is a challenging statistical task. In particular, we are interested in testing (1) whether the smooth ACTH circadian rhythms in chronic fatigue syndrome and fibromyalgia patients differ from those in healthy controls and (2) whether the patterns of pulsatile activities are different. In this paper, a hierarchical state space model is proposed to extract these signals from noisy observations. The smooth circadian rhythms shared by a group of subjects are modeled by periodic smoothing splines. The subject level pulsatile activities are modeled by autoregressive processes. A functional random effect is adopted at the pair level to account for the matched pair design. Parameters are estimated by maximizing the marginal likelihood. Signals are extracted as posterior means. Computationally efficient Kalman filter algorithms are adopted for implementation. Application of the proposed model reveals that the smooth circadian rhythms are similar in the two groups but the pulsatile activities in patients are weaker than those in the healthy controls.

Original languageEnglish (US)
Pages (from-to)3223-3234
Number of pages12
JournalStatistics in Medicine
Volume34
Issue number24
DOIs
StatePublished - Oct 30 2015

Fingerprint

Space Simulation
Circadian Rhythm
State-space Model
Hierarchical Model
Hormones
Modeling
Adrenocorticotropic Hormone
Posterior Mean
Matched pairs
Marginal Likelihood
Chronic Fatigue Syndrome
Smoothing Splines
Autoregressive Process
Random Effects
Fatigue
Kalman Filter
Testing
Profile
Evaluate

Keywords

  • Hierarchical models
  • Hormone profiles
  • Longitudinal data
  • Signal extraction
  • State space models

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

Modeling diurnal hormone profiles by hierarchical state space models. / Liu, Ziyue; Guo, Wensheng.

In: Statistics in Medicine, Vol. 34, No. 24, 30.10.2015, p. 3223-3234.

Research output: Contribution to journalArticle

@article{a0f19c01779045788a64bb4a9841a307,
title = "Modeling diurnal hormone profiles by hierarchical state space models",
abstract = "Adrenocorticotropic hormone (ACTH) diurnal patterns contain both smooth circadian rhythms and pulsatile activities. How to evaluate and compare them between different groups is a challenging statistical task. In particular, we are interested in testing (1) whether the smooth ACTH circadian rhythms in chronic fatigue syndrome and fibromyalgia patients differ from those in healthy controls and (2) whether the patterns of pulsatile activities are different. In this paper, a hierarchical state space model is proposed to extract these signals from noisy observations. The smooth circadian rhythms shared by a group of subjects are modeled by periodic smoothing splines. The subject level pulsatile activities are modeled by autoregressive processes. A functional random effect is adopted at the pair level to account for the matched pair design. Parameters are estimated by maximizing the marginal likelihood. Signals are extracted as posterior means. Computationally efficient Kalman filter algorithms are adopted for implementation. Application of the proposed model reveals that the smooth circadian rhythms are similar in the two groups but the pulsatile activities in patients are weaker than those in the healthy controls.",
keywords = "Hierarchical models, Hormone profiles, Longitudinal data, Signal extraction, State space models",
author = "Ziyue Liu and Wensheng Guo",
year = "2015",
month = "10",
day = "30",
doi = "10.1002/sim.6579",
language = "English (US)",
volume = "34",
pages = "3223--3234",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "24",

}

TY - JOUR

T1 - Modeling diurnal hormone profiles by hierarchical state space models

AU - Liu, Ziyue

AU - Guo, Wensheng

PY - 2015/10/30

Y1 - 2015/10/30

N2 - Adrenocorticotropic hormone (ACTH) diurnal patterns contain both smooth circadian rhythms and pulsatile activities. How to evaluate and compare them between different groups is a challenging statistical task. In particular, we are interested in testing (1) whether the smooth ACTH circadian rhythms in chronic fatigue syndrome and fibromyalgia patients differ from those in healthy controls and (2) whether the patterns of pulsatile activities are different. In this paper, a hierarchical state space model is proposed to extract these signals from noisy observations. The smooth circadian rhythms shared by a group of subjects are modeled by periodic smoothing splines. The subject level pulsatile activities are modeled by autoregressive processes. A functional random effect is adopted at the pair level to account for the matched pair design. Parameters are estimated by maximizing the marginal likelihood. Signals are extracted as posterior means. Computationally efficient Kalman filter algorithms are adopted for implementation. Application of the proposed model reveals that the smooth circadian rhythms are similar in the two groups but the pulsatile activities in patients are weaker than those in the healthy controls.

AB - Adrenocorticotropic hormone (ACTH) diurnal patterns contain both smooth circadian rhythms and pulsatile activities. How to evaluate and compare them between different groups is a challenging statistical task. In particular, we are interested in testing (1) whether the smooth ACTH circadian rhythms in chronic fatigue syndrome and fibromyalgia patients differ from those in healthy controls and (2) whether the patterns of pulsatile activities are different. In this paper, a hierarchical state space model is proposed to extract these signals from noisy observations. The smooth circadian rhythms shared by a group of subjects are modeled by periodic smoothing splines. The subject level pulsatile activities are modeled by autoregressive processes. A functional random effect is adopted at the pair level to account for the matched pair design. Parameters are estimated by maximizing the marginal likelihood. Signals are extracted as posterior means. Computationally efficient Kalman filter algorithms are adopted for implementation. Application of the proposed model reveals that the smooth circadian rhythms are similar in the two groups but the pulsatile activities in patients are weaker than those in the healthy controls.

KW - Hierarchical models

KW - Hormone profiles

KW - Longitudinal data

KW - Signal extraction

KW - State space models

UR - http://www.scopus.com/inward/record.url?scp=84942827857&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84942827857&partnerID=8YFLogxK

U2 - 10.1002/sim.6579

DO - 10.1002/sim.6579

M3 - Article

C2 - 26152819

AN - SCOPUS:84942827857

VL - 34

SP - 3223

EP - 3234

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 24

ER -