Maintaining homeostasis, the regulation of internal physiological parameters, is essential for health and well-being. Deviations from optimal levels, or ‘sweet spots,’ can lead to health deterioration and disease. Identifying biomarkers with sweet spots requires both change-point detection and variance effect analysis. Traditional approaches involve separate tests for change-points and heteroskedasticity, which can yield inaccurate results if model assumptions are violated. To address these challenges, we propose a unified approach: Bayesian Testing for Heteroskedasticity and Sweet Spots (BTHS). This framework integrates sampling-based parameter estimation and Bayes factor computation to enhance change-point detection, heteroskedasticity quantification, and testing in change-point regression settings, and extends previous Bayesian approaches. BTHS eliminates the need for separate analyses and provides detailed insights into both the magnitude and shape of heteroskedasticity, enabling robust identification of sweet spots without strong assumptions. We applied BTHS to blood elements from the Canadian Longitudinal Study on Aging identifying nine blood elements with significant sweet spot variance effects.
Get full access to this article
View all access options for this article.
References
1.
AmanY, Schmauck-MedinaT, HansenM, et al.Autophagy in healthy aging and disease. Nat Aging, 2021; 1(8):634–650.
2.
AmreinK, QuraishiSA, LitonjuaAA, et al.Evidence for a u-shaped relationship between prehospital vitamin d status and mortality: A Cohort Study. J Clin Endocrinol Metab, 2014; 99(4):1461–1469.
3.
AnsariF, GrayK, NathwaniD, et al.Outcomes of an intervention to improve hospital antibiotic prescribing: Interrupted time series with segmented regression analysis. J Antimicrob Chemother, 2003; 52(5):842–848.
4.
AnselminoM, WallanderM, NorhammarA, et al.Implications of abnormal glucose metabolism in patients with coronary artery disease. Diab Vasc Dis Res, 2008; 5(4):285–290.
5.
AyrolesJF, BuchananSM, O’LearyC, et al.Behavioral idiosyncrasy reveals genetic control of phenotypic variability. Proc Natl Acad Sci U S A, 2015; 112(21):6706–6711.
6.
BreuschTS, PaganAR. A simple test for heteroscedasticity and random coefficient variation. Econometrica, 1979; 47(5):1287–1294.
7.
BrunhamLR, RuelI, AljenedilS, et al.Canadian cardiovascular society position statement on familial hypercholesterolemia: Update 2018. Can J Cardiol, 2018; 34(12):1553–1563.
8.
ChenCW, ChanJS, GerlachR, et al.A comparison of estimators for regression models with change points. Stat Comput, 2011; 21(3):395–414.
9.
ChowGC. Tests of equality between sets of coefficients in two linear regressions. Econometrica, 1960; 28(3):591–605.
10.
CookRD, WeisbergS. Diagnostics for heteroscedasticity in regression. Biometrika, 1983; 70(1):1–10.
11.
DelanayeP, SchaeffnerE, EbertN, et al.Normal reference values for glomerular filtration rate: What do we really know? Nephrol Dial Transplant, 2012; 27(7):2664–2672.
12.
DickeyJM, LientzB. The weighted likelihood ratio, sharp hypotheses about chances, the order of a Markov chain. Ann Math Statist, 1970; 41(1):214–226.
13.
DrakeMT. Vitamin D and the Goldilocks principle: Too little, too much, or just right? J Clin Endocrinol Metab, 2014; 99(4):1164–1166.
14.
DuaneS, KennedyAD, PendletonBJ, et al.Hybrid Monte Carlo. Physics Letters B, 1987; 195(2):216–222.
15.
DumitrascuB, DarnellG, AyrolesJ, et al.Statistical tests for detecting variance effects in quantitative trait studies. Bioinformatics, 2019; 35(2):200–210.
16.
ErdmanC, EmersonJW. bcp: An r package for performing a bayesian analysis of change point problems. Journal of Statistical Software, 2008; 23:1–13.
17.
FedarkoNS. The biology of aging and frailty. Clin Geriatr Med, 2011; 27(1):27–37.
18.
GemanS, GemanD. Stochastic relaxation, gibbs distributions, and the bayesian restoration of images. IEEE Trans Pattern Anal Mach Intell, 1984; 6(6):721–741.
GoodrichBK, WawroG, KatznelsonI. Designing quantitative historical social inquiry: An introduction to stan. In; APSA 2012 Annual Meeting Paper, 2012.
21.
GronauQF, SarafoglouA, MatzkeD, et al.A tutorial on bridge sampling. J Math Psychol, 2017; 81:80–97.
22.
GronauQF, HeathcoteA, MatzkeD. Computing bayes factors for evidence-accumulation models using warp-iii bridge sampling. Behav Res Methods, 2020a;52(2):918–937.
23.
GronauQF, SingmannH, WagenmakersE-J. bridgesampling: An R package for estimating normalizing constants. J Stat Soft, 2020b;92(10):1–29; doi: 10.18637/jss.v092.i10
24.
GrundySM, BeckerD, ClarkLT, et al.Detection, evaluation, and treatment of high blood cholesterol in adults (adult treatment panel iii). Circulation, 2002; 106(25):3143–3421.
25.
GrundySM, CleemanJI, MerzCNB, et al.American Heart Association. Implications of recent clinical trials for the national cholesterol education program adult treatment panel iii guidelines. Circulation, 2004; 110(2):227–239.
26.
HeckDW. A caveat on the Savage-Dickey density ratio: The case of computing bayes factors for regression parameters. Br J Math Stat Psychol, 2019; 72(2):316–333.
27.
HoffmanMD, GelmanA, et al.The no-u-turn sampler: Adaptively setting path lengths in hamiltonian monte carlo. J Mach Learn Res, 2014; 15(1):1593–1623.
28.
HoffmanR, BenzEJ, SilbersteinLE, et al.Hematology: Basic Principles and Practice. Elsevier Health Sciences; 2022.
29.
HonkelaA. Computational statistics i, 2020.
30.
JacksonH, CarterK, DarkeC, et al.HFE mutations, iron deficiency and overload in 10 500 blood donors. Br J Haematol, 2001; 114(2):474–484.
31.
JeffreysH. The Theory of Probability. OuP Oxford, 1998.
32.
JohnsonJD, AlejandroEU. Control of pancreatic β-cell fate by insulin signaling: The sweet spot hypothesis. Cell Cycle, 2008; 7(10):1343–1347.
33.
KassRE, RafteryAE. Bayes factors. J Am Stat Assoc, 1995; 90(430):773–795.
34.
KassRE, VaidyanathanSK. Approximate bayes factors and orthogonal parameters, with application to testing equality of two binomial proportions. J R Stat Soc Series B Stat Methodol, 1992; 54(1):129–144.
35.
KaufmanRL. Heteroskedasticity in Regression: Detection and Correction. Sage Publications; 2013.
36.
KillickR, EckleyIA. Changepoint: An r package for changepoint analysis. J Stat Soft, 2014; 58(3):1–19.
37.
KimEH, MorseA, ZingalesL. What has mattered to economics since 1970. J Econ Perspect, 2006; 20(4):189–202.
38.
KoenkerR. A note on studentizing a test for heteroscedasticity. J Econom, 1981; 17(1):107–112.
39.
KorkasK, FryzlewiczP. Multiple change-point detection for non-stationary time series using wild binary segmentation. Stat SINICA, 2017:287–311.
40.
LeeJH, HanK, HuhJH. The sweet spot: Fasting glucose, cardiovascular disease, and mortality in older adults with diabetes: A nationwide population-based study. Cardiovasc Diabetol, 2020; 19(1):44.
41.
LewisSM, RafteryAE. Estimating bayes factors via posterior simulation with the Laplace-Metropolis estimator. J Am Stat Assoc, 1997; 92(438):648–655.
42.
LinP-H, DuannP. Dyslipidemia in kidney disorders: Perspectives on mitochondria homeostasis and therapeutic opportunities. Front Physiol, 2020; 11:1050.
43.
LindeløvJK, et al.mcp: An r package for regression with multiple change points. OSF Preprints, 2020; 10.
44.
MattesonDS, JamesNA. A nonparametric approach for multiple change point analysis of multivariate data. J Am Stat Assoc, 2014; 109(505):334–345.
NealRM, et al.MCMC using Hamiltonian dynamics. Handbook of Markov Chain Monte Carlo, 2011; 2(11):2.
55.
NishioM, ArakawaA. Performance of hamiltonian monte carlo and no-u-turn sampler for estimating genetic parameters and breeding values. Genet Sel Evol, 2019; 51(1):1–12.
56.
Pérez-ZepedaMU, GodinJ, ArmstrongJJ, et al.Frailty among middle-aged and older canadians: Population norms for the frailty index using the canadian longitudinal study on aging. Age Ageing, 2021; 50(2):447–456.
57.
PetrakisE, StamboliadisE, KomnitsasK. Evaluation of the relationship between energy input and particle size distribution in communication with the use of piecewise regression analysis. Particulate Sci Technol, 2017; 35(4):479–489.
58.
PomattoLC, SunPY, DaviesKJ. To adapt or not to adapt: Consequences of declining adaptive homeostasis and proteostasis with age. Mech Ageing Dev, 2019; 177:80–87.
59.
PunBT, BadenesR, Heras La CalleG, et al.COVID-19 Intensive Care International Study Group. Prevalence and risk factors for delirium in critically ill patients with covid-19 (covid-d): A multicentre cohort study. Lancet Respir Med, 2021; 9(3):239–250.
60.
QuandtRE. The estimation of the parameters of a linear regression system obeying two separate regimes. J Am Stat Assoc, 1958; 53(284):873–880.
61.
QuandtRE. Tests of the hypothesis that a linear regression system obeys two separate regimes. J Am Stat Assoc, 1960; 55(290):324–330.
62.
QueitschC, SangsterTA, LindquistS. HSP90 as a capacitor of phenotypic variation. Nature, 2002; 417(6889):618–624.
63.
RainaP, WolfsonC, KirklandS, et al.Cohort profile: The Canadian Longitudinal Study on Aging (CLSA). International Journal of Epidemiology, 2019; 48(6):1752–1753j.
64.
SchillingsC, SprungkB, WackerP. On the convergence of the Laplace approximation and noise-level-robustness of Laplace-based Monte Carlo methods for bayesian inverse problems. Numer Math, 2020; 145(4):915–971.
65.
SearleSD, MitnitskiA, GahbauerEA, et al.A standard procedure for creating a frailty index. BMC Geriatr, 2008; 8(1):24.
66.
SenapatiA, NagA, MondalA, et al.A novel framework for COVID-19 case prediction through piecewise regression in india. Int J Inf Technol, 2021; 13(1):41–48.
67.
Stan Development Team. RStan: The R interface to Stan. 2023. Available from: http://mc-stan.org/ R package version 2.21.8.
68.
StockdaleJE, DoigR, MinJ, et al.C. Long time frames to detect the impact of changing COVID-19 measures, Canada, March to July 2020. Eurosurveillance, 2021; 26(40):2001204.
69.
TomsJD, LesperanceML. Piecewise regression: a tool for identifying ecological thresholds. Ecology, 2003; 84(8):2034–2041.
70.
VishnyakovaO, XiaoweiS, KennethR, et al.Physiological phenotypes have optimal values relevant to healthy aging: Sweet spots deduced from the Canadian Longitudinal Study on Aging. Geroscience, 2024; 46(2):1589–1605.
71.
WernerR, ValevD, DanovD, et al.Study of structural break points in global and hemispheric temperature series by piecewise regression. Advances in Space Res, 2015; 56(11):2323–2334.
72.
WhiteH. A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica, 1980; 48(4):817–838.
73.
WilliamsMN, BååthRA, PhilippMC. Using Bayes factors to test hypotheses in developmental research. Res Human Develop, 2017; 14(4):321–337.
74.
WolfsonDW, AndersenDE, FiebergJR. Using piecewise regression to identify biological phenomena in biotelemetry datasets. J Anim Ecol, 2022; 91(9):1755–1769.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
