The estimation of population parameters using complex survey data requires careful statistical modelling to account for the design features. This is further complicated by unit and item nonresponse for which a number of methods have been developed in order to reduce estimation bias. In this paper, we address some issues that arise when the target of the inference (i.e. the analysis model or model of interest) is the conditional quantile of a continuous outcome. Survey design variables are duly included in the analysis and a bootstrap variance estimation approach is proposed. Missing data are multiply imputed by means of chained equations. In particular, imputation of continuous variables is based on their empirical distribution, conditional on all other variables in the analysis. This method preserves the distributional relationships in the data, including conditional skewness and kurtosis, and successfully handles bounded outcomes. Our motivating study concerns the analysis of birthweight determinants in a large UK-based cohort of children. A novel finding on the parental conflict theory is reported. R code implementing these procedures is provided.
ThompsonPCaiYZMoyeedR. Bayesian nonparametric quantile regression using splines. Comput Stat Data Anal2010; 54: 1138–1150.
11.
HeagertyPJPepeMS. Semiparametric estimation of regression quantiles with application to standardizing weight for height and age in US children. J R Stat Soc Ser C1999; 48: 533–551.
12.
LeeDNeocleousT. Bayesian quantile regression for count data with application to environmental epidemiology. J R Stat Soc Ser C2010; 59: 905–920.
13.
KoenkerRGelingO. Reappraising medfly longevity. J Am Stat Assoc2001; 96: 458–468.
14.
KoenkerR. Quantile regression, New York: Cambridge University Press, 2005.
15.
AbrevayaJ. The effects of demographics and maternal behavior on the distribution of birth outcomes. Empir Econ2001; 26: 247–257.
16.
AbrevayaJDahlCM. The effects of birth inputs on birthweight: evidence from quantile estimation on panel data. J Bus Econ Stat2008; 26: 379–397.
BurgetteLFReiterJPMirandaML. Exploratory quantile regression with many covariates: an application to adverse birth outcomes. Epidemiology2011; 22: 859–866.
19.
MuddLMPivarnikJHolzmanCB. Leisure-time physical activity in pregnancy and the birth weight distribution: where is the effect?J Phys Act Health2012; 9: 1168–1177.
20.
BörnhorstCHenseSAhrensW. From sleep duration to childhood obesity – what are the pathways?Eur J Pediatr2012; 171: 1029–1038.
21.
WeiYHeXM. Conditional growth charts (with Discussion). Ann Stat2006; 34: 2069–2097.
22.
WeiYPereAKoenkerR. Quantile regression methods for reference growth charts. Stat Med2006; 25: 1369–1382.
23.
ChenKMüllerHG. Conditional quantile analysis when covariates are functions, with application to growth data. J R Stat Soc Ser B2012; 74: 67–89.
24.
BeyerleinAToschkeAMvon KriesR. Breastfeeding and childhood obesity: shift of the entire BMI distribution or only the upper parts?Obesity2008; 16: 2730–2733.
25.
FenskeNKneibTHothornT. Identifying risk factors for severe childhood malnutrition by boosting additive quantile regression. J Am Stat Assoc2011; 106: 494–510.
26.
BirchJMGeraciMAlstonRD. Low birthweight and aetiology of childhood liver tumours in North West England. Pediatr Blood Cancer2010; 55: 932–932.
27.
BayerONeuhauserHVon KriesR. Sleep duration and blood pressure in children: a cross-sectional study. J Hypertension2009; 27: 1789–1793.
WoodruffRS. Confidence intervals for medians and other position measures. J Am Stat Assoc1952; 47: 635–646.
30.
SedranskJMeyerJ. Confidence intervals for the quantiles of a finite population: simple random and stratified simple random sampling. J R Stat Soc Ser B1978; 40: 239–252.
31.
ShaoJ. L-statistics in complex survey problems. Ann Stat1994; 22: 946–967.
32.
ChatterjeeA. Asymptotic properties of sample quantiles from a finite population. Ann Inst Stat Math2011; 63: 157–179.
33.
WangJCQOpsomerJD. On asymptotic normality and variance estimation for nondifferentiable survey estimators. Biometrika2011; 98: 91–106.
34.
GeraciMSalvatiN. The geographical distribution of the consumption expenditure in Ecuador: estimation and mapping of the regression quantiles. Stat Appl – Ital J Appl Stat2007; 19: 167–183.
35.
LiYGraubardBIKornEL. Application of nonparametric quantile regression to body mass index percentile curves from survey data. Stat Med2010; 29: 558–572.
36.
HeYZaslavskyALandrumM. Multiple imputation in a large-scale complex survey: a practical guide. Stat Meth Med Res2010; 19: 653–670.
Van BuurenS. Multiple imputation of discrete and continuous data by fully conditional specification. Stat Meth Med Res2007; 16: 219–242.
39.
BottaiMZhenH. Multiple imputation based on conditional quantile estimation. Epidemiol Biostat Pub Health2013; 19(1)): . DOI: 10.2427/8758.
40.
ChengPEChuCK. Kernel estimation of distribution functions and quantiles with missing data. Stat Sin1996; 6: 63–78.
41.
LipsitzSRFitzmauriceGMMolenberghsG. Quantile regression methods for longitudinal data with drop-outs: application to CD4 cell counts of patients infected with the human immunodeficiency virus. J R Stat Soc Ser C1997; 46: 463–476.
42.
YuanYYinG. Bayesian quantile regression for longitudinal studies with nonignorable missing data. Biometrics2010; 66: 105–114.
43.
WeiYMaYCarrollRJ. Multiple imputation in quantile regression. Biometrika2012; 99: 423–438.
44.
R Development Core Team. R: a language and environment for statistical computing, Vienna, Austria: R Foundation for Statistical Computing, 2011.
45.
LumleyT. Analysis of complex survey samples. J Stat Software2004; 9: 1–19.
46.
LumleyT. Survey: analysis of complex survey samples. R package version 3.24-1. 2011.
47.
KoenkerR. quantreg: quantile regression. R package version 4.65. 2011.
48.
Van BuurenSGroothuis-OudshoornK. mice: multivariate imputation by chained equations in R. J Stat Software2011; 45: 1–67.
49.
PlewisI. The Millennium Cohort Study: technical report on sampling, 4th ed. London: Centre for Longitudinal Studies, 2007.
50.
CullisA. Infant mortality in the Millennium Cohort Study (MCS) sample areas, London: Centre for Longitudinal Studies, 2004.
51.
PlewisI. Non-response in a birth cohort study: the case of the Millennium Cohort Study. Int J Soc Res Methodol2007; 10: 325–334.
52.
KoenkerR. Quantile regression for longitudinal data. J Multivar Data Anal2004; 91: 74–89.
53.
GeraciMBottaiM. Quantile regression for longitudinal data using the asymmetric Laplace distribution. Biostatistics2007; 8: 140–154.
54.
GeraciMBottaiM. Linear quantile mixed models. Stat Comput2013. DOI: 10.1007/s11222-013-9381-9.
55.
GeraciM. lqmm: linear quantile mixed models. R package version 1.02. 2012.
56.
PortnoySKoenkerR. The Gaussian hare and the Laplacian tortoise: computability of squared-error versus absolute-error estimators. Stat Sci1997; 12: 279–300.
57.
CantyAJDavisonAC. Resampling-based variance estimation for labour force surveys. J R Stat Soc Ser D1999; 48: 379–391.
58.
PresnellBBoothJG. Resampling methods for sample survey, Florida: Department of Statistics, University of Florida, 1994.
59.
DavisonACHinkleyDV. Bootstrap methods and their application, Cambridge, UK: Cambridge University Press, 1997.
ChernozhukovVFernández-ValI. Inference for extremal conditional quantile models, with an application to market and birthweight risks. Rev Econ Stud2011; 78: 559–589.
62.
SchwartzSLGelfandAEMirandaML. Joint Bayesian analysis of birthweight and censored gestational age using finite mixture models. Stat Med2010; 29: 1710–1723.
63.
GriffithsLJDezateuxCColeTJ. Differential parental weight and height contributions to offspring birthweight and weight gain in infancy. Int J Epidemiol2007; 36: 104–107.
RubinDB. Multiple imputation for nonresponse in surveys, New York: Wiley, 1987.
66.
ReiterJPRaghunathanTEKinneySK. The importance of modeling the sampling design in multiple imputation for missing data. Surv Methodol2006; 32: 143–150.
67.
WhiteIRRoystonPWoodAM. Multiple imputation using chained equations: issues and guidance for practice. Stat Med2011; 30: 377–399.
68.
LittleRJA. Missing-data adjustments in large surveys. J Bus Econ Stat1988; 6: 287–296.
69.
SchenkerNTaylorJMG. Partially parametric techniques for multiple imputation. Comput Stat Data Anal1996; 22: 425–446.
70.
HeYRaghunathanTE. Tukey’s gh distribution for multiple imputation. Am Stat2006; 60: 251–256.
71.
RaghunathanTELepkowskiJMHoewykJV. A multivariate technique for multiply imputing missing values using a sequence of regression models. Surv Methodol2001; 27: 85–95.
72.
PortnoyS. Asymptotic behavior of the number of regression quantile breakpoints. SIAM J Sci Stat Comput1991; 12: 867–883.
73.
SuYSGelmanAHillJ. Multiple imputation with diagnostics (mi) in R: opening windows into the black box. J Stat Software2011; 45: 1–31.
GutenbrunnerCJurečkováJ. Regression rank scores and regression quantiles. Ann Stat1992; 20: 305–330.
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.
