1 Introduction
The remarkable works on anthropometry by Adolphe Quetelet and Sir Francis Galton in the 19th century gave birth to a new field of scientific investigation within which the medical and statistical sciences developed a longlasting and profitable collaboration. In turn, this has given rise to countless research studies in public health and to the development of important analytic methods. For example, the body mass index (BMI), also known as the Quetelet index
, is universally applied by researchers and clinicians to classify individuals into categories such as ‘underweight’, ‘overweight’, and ‘obese’ as these may be at higher risks of poorer health outcomes. Classification thresholds for these categories are defined as percentiles of the BMI distribution in a reference population and are published by public health institutes like the World Health Organization (
http://www.who.int/childgrowth/standards/en/).The BMI index is a wellknown example of ratio of anthropometric variables (mass/height). Another similar ratio is the corpulence or Rohrer index (mass/height). Indeed, there exists a plethora of indices where body measurements are combined as ratios, often upon power transformations, where the scaling depends on the relationship between the variables involved in the ratio. If the correct scaling exponent is applied, then no residual association should be observed between the index and the scaling variable (Heymsfield et al., 2007). Otherwise, differential misclassification bias may result when the goal is to assess risk in different categories of the anthropometric index (Diverse Populations Collaborative Group, 2005).
In this paper, we are specifically interested in anthropometric measurements for very preterm infants (22 to 29 weeks’ gestation). Preterm babies, particularly those born at lower gestational ages, have high risks of mortality, morbidities, and neurodevelopmental impairment (Stoll et al., 2010; Horbar et al., 2012)
. For example, it is estimated that at 22 and 23 weeks’ gestation the mortality rate can be as high as 80%
(Stoll et al., 2010). At these ages, there are significant rates of respiratory distress syndrome (94%), patent ductus arteriosus (46%), severe intraventricular hemorrhage (16%), necrotizing enterocolitis (11%), and lateonset sepsis (36%) (Stoll et al., 2010).Preterm birth is not the only risk factor. Growth restriction, usually defined as birthweight (BW) less than the 10th percentile for gestational age (GA)—or small for gestational age (SGA)—further raises already high risks among preterm infants (Bernstein et al., 2000) and, hence, is used as an indicator for secondary and tertiary prevention of mortality and adverse outcomes. The etiology of SGA is multifactorial with some causes linked to, for example, smoking, placental insufficiency, environmental factors, and maternal complications like preeclampsia. These factors not only impact BW but might also affect the size of the head (as measured by head circumference, HC, right after birth), with consequences that may vary according to the particular period of pregnancy in which the insult has occurred. It has been theorized that: if the insult occurs early during pregnancy or even before conception, growth restriction is symmetric (or proportional) and both BW and HC are affected; if the insult occurs later during pregnancy, growth restriction is asymmetric (or disproportional), with negative consequences mainly for BW, thus resulting in a larger HCtoBW ratio (Vandenbosche and Kirchner, 1998; Saleem et al., 2011). Asymmetric growth is thought to be an adaptive mechanism that is put in place to protect the brain. In response to placental insufficiency, which is often caused by hypertension and leads to intrauterine growth restriction, the fetus adapts its circulation to preserve oxygen and nutrient supply to the brain (the ‘brainsparing’ effect). Some studies investigated the determinants of fetal growth and body proportionality, as well the effect of the latter on neonatal outcomes. They found that (severe) pregnancyrelated hypertension, which is the development of new hypertension after 20 weeks’ gestation, is strongly associated with a larger HCtoBW ratio (Kramer et al., 1990a). The latter, in turn, was found to be a risk factor for stillbirth and fetal distress (Kramer et al., 1990b).
Thus, it is natural to wonder whether there are differences in terms of health outcomes between infants with unusual HCtoBW ratio and those with normal HCtoBW ratio. The abnormality of a ratio reflects the abnormality of either the numerator or denominator, relative to the other. This is illustrated in Figure 1 which shows the relationship between BW and HC in preterm infants using data from the Vermont Oxford Network (VON), a large network of medical centers. The observation marked by a thick cross in the left plot of Figure 1 (a girl born at 162 days of gestation) presents an abnormal HCtoBW ratio for her GA (right plot), a possible consequence of the brainsparing effect. Note that, individually, the BW and HC measurements for this girl do not present a concern since they both lie above the respective cutoffs for small BW and small HC.
The characterization of abnormal bivariate observations require the application of multivariate approaches to joint ranking. Multivariate modelling has a long tradition in parametric statistics. Models for joint distributions are, more or less, direct extensions of wellknown univariate distributions to higher dimensions. The multivariate normal distribution is, among others, often invoked for its mathematical and statistical properties. In the past few years, there has been a growing number of applications giving stronger attention to distributions that can flexibly account for heavier tails
(Goodman and Kotz, 1973), skewness
(Azzalini and DallaValle, 1996; Kozubowski and Podgorski, 2000; Kozubowski, Podgórski and Rychlik, 2013) and, more in general, to nonelliptical distributions. Rather than assuming a specific parametric distribution, we take a more agnostic approach and we propose to investigate these issues by using directional quantile envelopes (DQEs) for multivariate data (Kong and Mizera, 2012).The interpretation of a DQE is straightforward. In a bivariate plot, a DQE is represented by a contour line with constant quantile level. Each point on the contour line can be mapped to a fixed percentile of the distribution of the data coordinates’ projections onto the real line. The projections are obtained in every possible direction on the circle. The key idea is to divide observations in two groups: those that lie within the contour line (jointly normal) and those outside (jointly abnormal). However, in our specific application investigating the brain sparing effect, the set of all data points classified as jointly abnormal is unsatisfactory as it comprises a clinically heterogeneous mix of infants. In particular, it does not distinguish between infants with symmetric growth and those with asymmetric growth. Rather, we want to focus on infants that, on the one hand, have jointly abnormal measurements, but, on the other, have a large HCtoBW ratio. This is equivalent to choosing a directional quantile in a particular direction. We propose using allometry to determine such a direction.
Besides the study of Kong and Mizera (2012), there are other precedents of applications of methods for multivariate quantiles to anthropometric and growth charts. Some authors (McKeague et al., 2011) proposed quantile contours based on Tukey’s notion of halfspace depth, while others (Wei, 2008) considered directional reference intervals built around a central point of the distribution (location parameter). All these studies offer nonparametric approaches to identify jointly abnormal measurements when parametric (normality) assumptions are inappropriate. Therefore, we do not claim any novelty in this regard. In contrast, our focus is on the classification of abnormal ratios. Moreover, we bring forward a connection between DQEs (Kong and Mizera, 2012) and allometric modelling which, to our knowledge, does not seem to have been reported before. This in turn provides an operational definition of directional quantile that can be used as cutoff for risk assessment, as demonstrated in the analysis of the VON data.
The rest of the paper is organized as follows. In Section 2, we give a general overview of VON and some details about the variables of interest. In Section 3.1, we briefly discuss quantiles for univariate data and their limitations when used as individual cutoffs in multivariate problems. In Section 3.2, we provide the formal definition of DQE and introduce the relationship between directional quantiles and ratios. In Section 3.3, we establish the connection between DQEs and allometry. In Section 4, we apply the proposed methods to the VON data to investigate mortality risk in proportionately and disproportionately growthrestricted infants. We show that our principled approach stratifies mortality risk more effectively as compared to the commonlyadopted approach based on isometric scaling. Although our methods focus on anthropometry and allometry, extensions and generalizations to other settings are discussed in Section 5.
2 The Vermont Oxford Network
The Vermont Oxford Network (VON) is a nonprofit, voluntary collaboration of health care professionals ‘dedicated to improving the quality and safety of medical care for newborn infants and their families through a coordinated program of research, education, and quality improvement projects’ (https://public.vtoxford.org/aboutus/). The Network was established in 1988 and comprises over 1200 centers (hospitals) with a neonatal intensive care unit (NICU). The Very Low Birth Weight Database collects information from these centers which account for approximately of all the births occurring at 2229 weeks of gestation in the United States (US). A number of variables are collected from each center: maternal characteristics (e.g., ethnicity), infant characteristics (e.g., sex, gestational age, birthweight, head circumference, birth defects), and newborn health outcomes (inhospital mortality and major morbidities). Member hospitals collect the data using uniform definitions through medical record abstraction which are then submitted to VON electronically or through paper forms. Data pass automated checks and are returned for correction if needed.
Females  Males  

Gestational age  BW  HC  BW  HC  
(21,23]  5,496  550  20.5  6,036  580  21.0 
(23,25]  17,398  680  22.0  19,305  730  22.5 
(25,27]  22,448  890  24.0  24,996  950  24.5 
(27,29]  28,097  1,145  26.0  31,970  1,220  26.5 
All ages  73,439  880  24.0  82,307  945  24.5 
Our study sample was restricted to inborn, singleton US infants born at 22 to 29 weeks of gestation between 2006 and 2014, with no congenital malformations. GA was determined using obstetrical measures based on prenatal ultrasound (accuracy days), last menstrual period (accuracy days), or a neonatologist’s estimate based on postnatal physical examinations (accuracy days for Dubowitz examination). BW was recorded from labor and delivery or, if unavailable, upon admission to the neonatal unit. HC was recorded on the day of birth or the day after. The data underwent some mild cleaning procedures as described elsewhere (Boghossian et al., 2016). In particular, we excluded infants with missing information on vital status (841), unknown gender (30), missing (71) or implausible (744) BW, missing (24,706) or implausible (860) HC, missing hospital lengthofstay (78), or who were hospitalized for longer than one year (565). Overall, about infants were excluded, leaving 155,746 infants for our analysis. Table 1 gives summary statistics of the sample by sex and four gestational age intervals: (21,23], (23,25], (25,27], and (27,29] weeks. We have divided gestational age in intervals mainly for practical reasons as we want to provide a summary of the results which can be readily used by public health and clinical practitioners. Each gestational age group comprises infants whose mortality risk is comparable within that interval. The sample size of the first interval is noticeably smaller than the other three (Table 1), but this interval stands out because it includes infants with the highest risk of mortality (Boghossian et al., 2018a). These data were used in previous publications (Boghossian et al., 2016, 2018b) to generate BW and HCforgestationalage percentile charts for clinical use.
3 Quantilebased risk classification
3.1 Quantiles of univariate data
The quantile function (QF) of a random variable
with cumulative distribution function (CDF)
is defined as(1) 
Here we assume that
is absolutely continuous with probability density function
over the support of . Therefore the QF is simply the inverse of the CDF, .In the presence of covariates, the QF can be extended to conditional distributions. The linear specification of the QR model is (Koenker and Bassett, 1978)
(2) 
where is a
dimensional vector and
is a vector of coefficients indexed by the quantile level . A generalization of (2) defines(3) 
where the transformation can be modelled either parametrically or nonparametrically. Moreover, if is monotone, then , which we call transformation rule (Gilchrist, 2000) (also known as equivariance to monotone transformations). It is worth mentioning here that quantiles enjoy a number of other properties (Gilchrist, 2000), including the reflection rule .
In clinical settings, it is customary to define a cutoff for abnormal measurements. We will use the terms ‘normal’, ‘subnormal’ (below normal), ‘supranormal’ (above normal), and ‘abnormal’ (not normal, either below or above) to classify observations based on arbitrary cutoffs but without giving them any clinical or diagnostic connotation. Cutoffs are often related to specific quantiles of the distribution. For example, infants are classified as SGA if their BW is below the 10th percentile of the BW distribution conditional on gestational age; otherwise, they are termed appropriate for gestational age (AGA). Assuming a model as in (3), the cutoff would be determined as
for some suitable transformation (Geraci and Jones, 2015; Boghossian et al., 2016). Similarly, infants are said to have a subnormal head size if their HC is below the 10th percentile of the HC distribution conditional on GA, that is
assuming, as before, that model (3) holds for some transformation (Geraci and Jones, 2015; Boghossian et al., 2016). Clearly, these cutoffs need be estimated, either externally using a representative sample from the standard or referent population of interest, or internally from within the same data.
Most of the times, BW and HC are treated separately in statistical analyses. For example, they are analyzed as separate outcomes or as ‘independent’ predictors of postnatal child outcomes. However, there are good reasons why it could be informative to analyze these variables jointly. First of all, BW and HC are necessarily related. Larger weights correspond to larger head circumferences, although the younger the baby the larger the head size in relation to the size of body. In other words, the HCtoBW ratio decreases with age (Figure 1). Secondly, health outcomes may differ among infants whose BW and HC are ranked jointly normal or abnormal, as those with symmetric and asymmetric growth. Joint ranking necessitates multivariate approaches.
3.2 Quantiles of multivariate data
Let denote a multivariate random vector collecting measurements for continuous variables (e.g., BW and HC) and let be a normalized (with unit norm) direction of dimension . The th directional quantile, in the direction , is the th quantile of the corresponding projection of the distribution of , that is . The supporting halfspace determined by is . The th directional quantile envelope (DQE) generated by is given by the intersection (Kong and Mizera, 2012)
(4) 
In a bivariate space (), the geometric intuition behind (4) is as follows. Consider a scatter of points as that on the left plot in Figure 1; fix equal to 0.1; and define a direction on the circle (for example, the westeast or southnorth direction). Next, we cumulate data points while moving along the chosen direction and we stop when the cumulative proportion is . We demarcate a line which divides the plane into two halfplanes, a ‘lower’ halfplane with of the points, and an ‘upper’ halfplane with the remaining . If we repeat this process for all the possible directions on the circle, then the intersection of all the demarcation lines defines an ovalshaped contour within which the data points belong to the upper halfplanes in all directions (an illustration is given in Figure 2). These data points represent the set . Similarly, the points outside the perimeter, which we denote by , belong to the lower halfplanes in some directions.
From the above exemplification it becomes clear that the proportion of points that are in will be less than since these points satisfy for all . It also becomes clear that the quantile in a given direction is equivalent to the quantile in the opposite direction, e.g., the th directional percentile in the southnorth direction is equivalent to the th directional percentile in the northsouth direction (hence, ).
Directional quantiles can be easily extended to conditional distributions. If we assume a linear model as in (2), we obtain
(5) 
Then the DQE (4) can be applied to the conditional quantiles in (5).
Model (5) presupposes the additivity of the coordinates of since . If, instead, a multiplicative relationship is to be studied, then the logarithmic transformation of the measurements is more appropriate (Kong and Mizera, 2012), i.e.
(6) 
where . The th directional quantiles of the coordinates on the logscale therefore corresponds to the th quantile of the logratio of the scaled coordinates, that is
(7) 
The ratios of the type are ubiquitous in public health and clinical applications. One can immediately recognize that is the BMI index if is height, is weight, and . The determination of this particular direction might seem somewhat axiomatic but, far from it, it is the result of statistical observations by Quetelet himself and, some 130 years later, Ancel Keys. However, Quetelet in his treatise (Quetelet, 1842) had already recognized that the choice of scaling was rather complex:
If man increased equally in all dimensions, his weight at different ages would be as the cube of his height. Now, this is not what we really observe. The increase of weight is slower, except during the first year after birth; […] However, if we compare two individuals […] we shall find that the weight of developed persons, of different heights, is nearly as the square of the stature.
Allometry studies the geometrical relationships in the human body. In the following section, we discuss the connection between allometry and DQEs. Such a connection provides an operational definition of directional quantile that can be used as cutoff for risk assessment.
3.3 Bivariate percentiles and allometric analysis
Let us consider an allometric model of the type
(8) 
. Equation (8) implies that the allometric ratio is constant and equal to .
In our specific application where is BW and is HC, the scaling exponent captures the differential growth ratio between the head and the body as a whole. For , the variables and are said to be isometric. Model (8), if correctly specified, provides a benchmark against which we can classify abnormal HCtoBW ratios. This is asserted in the following proposition which establishes the connection between (4) and (8).
Proposition 1.
Let and be two continuous and strictly positive random variables, and assume that the allometric model , , holds true. Also, define and assume that the th directional quantile envelope generated by is smooth. Then the lines
(9) 
where , are tangent to .
Proof.
We only need to prove that the line tangent to is of the form (1). By definition of , the former is given by
for any given direction .
On the logscale, the allometric equation given in Proposition 1 relating to can be rewritten as
(10) 
which has the same form of (7) with and . Therefore, for , which we call allometric direction, the line tangent to is . Now, by (7) we have that . Since the logarithm is a monotone transformation, we use the transformation rule introduced in Section 3.1 and obtain . Then the tangent line equation becomes , which corresponds to the first equation given in (1). To obtain the second equation in (1), it is sufficient to notice that , where the first equality follows from the reflection rule and the second equality follows from the transformation rule.
For a rigorous proof of the geometric properties of , the reader is referred to Kong and Mizera (2012). ∎
Two corollaries to Proposition 1 follow.
Corollary 1.
The tangent lines are unique.
Corollary 2.
The tangent halfspaces are the sets of points such that and .
The first corollary follows from the smoothness of . The second corollary is a consequence of the definition of . More importantly, this corollary provides the operational definition of subnormal () and supranormal () ratios corresponding to the allometric direction. The dashed line in the right plot of Figure 2 gives an illustration of the allometric direction. One may wonder if there is anything special about this direction. The answer lies in the properties of the estimator of . In particular, if
is estimated using MA regression, the directional quantile in the allometric direction is in the same direction as the principal axis of the bivariate normal ellipse fitted to the logtransformed data. It is wellknown that this is the direction of the first eigenvector of the variancecovariance matrix of
. Of course, there is nothing necessarily prescriptive about the normal distribution, so one can explore an alternative estimator for under a different distribution, if that distribution has a theoretical or empirical relevance, or use a nonparametric estimator. For example, it is common to estimate the slope of MA regression under the assumption of homoscedasticity for the logadditive counterpart of model (8). If necessary, this assumption may be relaxed to improve on accuracy and efficiency of the estimates by introducing a variance function of the type , where and . Alternatively, using a distributionfree approach, one could estimate by means of median regression on the logscale (Geraci, Alston and Birch, 2013), with advantages in terms of robustness to outliers and error distribution, as well as in terms of lossless transformation between scales due to the equivariance property. Moreover, such estimator has a close relationship with the Laplace distribution. However, the median estimator would not be robust to measurement error in the covariates and hence would require the application of methods that are computationally more complex than MA regression
(Wei and Carroll, 2009; Mao, Wei and Liu, 2017).In general, there is a stronger motivation for using the allometric direction and this is related to body proportionality in human growth assessment. If we consider studies on infants, the definition of ‘proportionality’ varies from study to study, where body weight is sometimes related to HC, or abdomen circumference, or more commonly to length. Except for some studies (e.g., as those based on the Rohrer index (Olsen et al., 2009)), several other implicitly assume isometric scaling by defining ratios of the type (Kramer et al., 1990a; Lin, Su and River, 1991; Williams and O’Brien, 1998; Dashe et al., 2000). However, if the true scaling exponent is different from one, then the ratio
depends on . This requires that a definition of abnormal ratio should be based on the conditional quantile , not on the marginal . The consequence of using the latter would be a misclassification of infants in categories of possibly different risks, which is the source of differential misclassification bias (Heymsfield et al., 2007; Diverse Populations Collaborative Group, 2005) we referred to in Section 1. We provide empirical evidence that misclassification bias and, in turn, poorer risk classification result in the analysis of the VON data when is arbitrarily fixed equal to 1 (Section 4.2).
Our discussion so far has focused on two variables only, mainly because we are interested in asymmetric growth restriction which is commonly defined using BW and HC only (Lin, Su and River, 1991; BoccaTjeertes et al., 2014; Guellec et al., 2015), but also because the VON data do not provide anthropometric variables besides BW and HC. However, it is natural to consider a generalization for . One of the first problems we would encounter is, obviously, visualizing a multivariate DQE in more than, say, three or four dimensions. However, computationally (4) can be applied for any . In contrast, equation (8) does not seem to have an immediate multivariate counterpart and different approaches can be considered. A popular approach to multivariate allometry is based on principal component analysis (Jolicoeur, 1963; Corruccini, 1983) (PCA) due to its geometric properties. In Appendix A, we provide a generalization of Proposition 1 using PCA to sketch the main idea.
4 Mortality risk in preterm infants
4.1 Risk classification based on univariate percentiles
As discussed in Section 1
, SGA infants and infants with subnormal HC are at increased risk of poor health outcomes. We can investigate exposureoutcome associations using appropriate regression models. In particular, if we have both BW and HC as exposures, we can define a categorical variable
with categories ‘normal BW, normal HC’, ‘subnormal BW, normal HC’, ‘subnormal BW, subnormal HC’, and ‘normal BW, subnormal HC’. In Figure 1, these categories correspond to points in quadrants I, II, III, and IV, respectively. Let be the outcome of interest (death) and a vector of covariates associated with (sex, gestational age, and their interaction). Using the VON data, we fitted the generalized linear modelwith loglink function .
Table 2 shows the gestationalageadjusted mortality risk of preterm infants born between 22 and 29 weeks. The baseline is given by infants with normal (i.e., above the 10th percentile for gestational age and sex) BW and HC. The baseline risk is and for girls and boys, respectively. The risk increases by (girls) or (boys) when only HC is subnormal and by about when only BW is subnormal. However, if both HC and BW are below their respective cutoffs, then the mortality risk is approximately 2.5 times the baseline risk, meaning that 1 out of 3 of these infants does not survive. The gestationalagespecific mortality risk for girls and boys is given in Appendix B (Tables 6 and 7, respectively). The relative risk for infants with subnormal BW and HC increases with gestational age, although the baseline risk of normal infants is highest at the lowest gestational ages.
BW  HC  Risk  Lower  Upper  

Females  
Normal  Normal  63,523  0.11  0.11  0.11 
Normal  10th  2,616  1.67  1.54  1.82 
10th  Normal  3,155  1.82  1.69  1.96 
10th  10th  4,145  2.57  2.44  2.71 
Males  
Normal  Normal  71,070  0.14  0.13  0.14 
Normal  10th  3,079  1.41  1.31  1.52 
10th  Normal  3,606  1.81  1.70  1.92 
10th  10th  4,552  2.43  2.32  2.54 
The risk categories in Table 2 are defined using separate rankings for BW and HC. This is the approach taken by some authors (Lin, Su and River, 1991; Guellec et al., 2015). However, this approach presents a difficulty. While the specific cutoff values (e.g., 10th percentile) may be relevant for BW and HC taken individually, nothing can be said about the joint ranking of the measurements. We have already noticed that the observation marked by a thick cross in the left plot of Figure 1 presents normal (for gestational age) weight (489 g) and HC (30 cm) and thus falls in the baseline group. Upon closer inspection, this infant has a rather extreme HCtoBW ratio (right plot in Figure 1), a possible consequence of the brainsparing effect.
In the next section, we examine the categories in Table 2 in more detail to see whether the abnormality of the HCtoBW ratio represents an additional risk factor. In particular, the categories ‘normal BW, normal HC’ and ‘subnormal BW, subnormal HC’ are of clinical interest since they represent the baseline and the highest mortality risk categories, respectively. The category ‘subnormal BW, normal HC’ comprises infants with a BWHC imbalance. Some studies use this imbalance to define asymmetric growth restriction (Guellec et al., 2015). However, as we previously argued, this definition is limited as it does not take into account the proportionality between BW and HC.
4.2 Risk classification based on the allometric direction
We first explore the bivariate distribution of BW and HC using DQEs conditional on gestational age. We use DQEs at level since the directional quantile coincides with the univariate 10th percentile for and for which are commonly used as clinical cutoffs.
We can take advantage of model (6) to estimate the DQE of at level conditional on gestational age, i.e.
where the reference is weeks and ,
, are dummy variables for the other gestational age intervals. The left plot of Figure
2 shows the estimated DQEs for girls in each age interval. The points that lie outside a particular contour line (i.e., the set ) represent BW and HC measurements that are jointly abnormal as compared to those that fall in , conditional on a specific gestational age group. We should note that, while informative, the classification based on the DQE is broad since the ‘abnormal’ labelling of measurements outside the DQE applies to disparate groups: infants with subnormal BW and HC, infants with supranormal BW and HC, and those with asymmetric BW and HC. We may want to focus on a particular group of measurements that are jointly abnormal, like those for which (supranormal HCtoBW ratio) that are suggestive of the brainsparing effect.We then fitted model (8) for BW and HC on the logscale using standardized major axis (MA) regression as implemented in the R package smatr (Warton et al., 2012). The reason for this choice lies in the likely presence of measurement error in both variables (Warton et al., 2006). Table 3
shows estimates of the coefficients and standard errors for all infants as well as by sex and by gestational age. Overall, the estimated coefficient
hints at the cubic relationship between length and volume, although the test of the null hypothesis
gave a value less than 0.001. The estimated scaling exponent was almost identical for boys and girls (), but changed significantly () across gestational ages. The rightmost column of Table 3 shows the th percentile of the estimated allometric ratio . An illustration for girls born at weeks is shown in the right plot of Figure 2. The dashed line marks the th percentile of the HCtoBW ratio in the allometric direction (note that this is tangent to the DQE at level ). Points to its left have a relatively large value of HC as compared to that of BW. Among these, marked by a thick cross, we find the girl born at 162 days of gestation that was featured in Figure 1. Without question, this observation seems to be rather extreme even at .Parameter  
All  0.4488 (0.0011)  0.3166 (0.0004)  2.9647 (0.0006) 
Sex (value )  
Females  0.4488 (0.0017)  0.3166 (0.0006)  2.9641 (0.0008) 
Males  0.4497 (0.0016)  0.3164 (0.0005)  2.9704 (0.0009) 
Gestational age (value )  
(21,23]  0.1816 (0.0087)  0.4125 (0.0032)  1.6123 (0.0018) 
(23,25]  0.4212 (0.0034)  0.3260 (0.0012)  2.7798 (0.0013) 
(25,27]  0.5102 (0.0027)  0.2963 (0.0009)  3.4026 (0.0011) 
(27,29]  0.5413 (0.0024)  0.2865 (0.0008)  3.6553 (0.0010) 
Using the results in Table 3, we can calculate the allometric ratio for each infant, where is the estimated coefficient for gestational age group , , and classify these ratios based on the gestationalagespecific cutoffs . Table 4 compares gestationalageadjusted mortality in infants with normal () and supranormal () ratios. In these two groups, the baseline mortality risk is comparable for infants with normal BW and HC: between 0.11 and 0.13 for females, and 0.14 for males. However, the mortality risk for infants with small BW and HC is about three times the baseline risk in those with supranormal ratios, but about twice the baseline risk in those with a normal ratio. As a consequence, absolute risks too differ greatly. For example, disproportionately small boys have an absolute risk of while proportionately small boys have an absolute risk of . Note that in Table 4 the category of infants with normal BW and subnormal HC has been omitted since, by definition, there are no infants with supranormal ratios and subnormal HC.
In summary, preterm infants that have small BW and HC (below their respective 10th percentiles) are at high risk of mortality, with a relative risk of around 2.5 as compared to infants with normal BW and HC (Table 2). However, this relative risk is an ‘average’ of a lower risk in proportionately small infants and a higher risk in disproportionately small infants (Table 4). Hence, the classification based on the allometric direction tangent to the DQE identifies the group of disproportionately growthrestricted infants as those with considerably high risk among small infants.
As we repeatedly mentioned, if the correct scaling exponent is not applied, then misclassification bias may result and distort risk assessment. In Appendix B (Table 8), we report estimates of mortality risk for infants with normal and supranormal ratios, where the ratio is calculated using the isometric scaling, i.e. . It is apparent that the mortality risk for infants with small BW and HC is not dissimilar from the baseline risk in those with supranormal ratios, while the magnitude of the estimated relative risk in those with normal ratios is noticeably smaller than that obtained in Table 4. Moreover, the classification based on the isometric scaling leads to relative risk estimates less than (although not significantly different from) 1 in one of the female groups, which is contrary to the wellestablished notion that SGA infants are at higher mortality risk.
BW  HC  Risk  Lower  Upper  Risk  Lower  Upper  

Females  Normal HCtoBW ratio  Supranormal HCtoBW ratio  
Normal  Normal  59,033  0.11  0.11  0.11  4,490  0.13  0.12  0.14 
10th  Normal  1,425  1.70  1.52  1.90  1,730  1.62  1.44  1.82 
10th  10th  3,284  2.30  2.16  2.45  861  3.14  2.82  3.50 
Males  Normal HCtoBW ratio  Supranormal HCtoBW ratio  
Normal  Normal  65,984  0.14  0.13  0.14  5,086  0.14  0.13  0.15 
10th  Normal  1,425  1.77  1.61  1.95  2,181  1.81  1.64  2.00 
10th  10th  3,384  2.16  2.04  2.28  1,168  3.17  2.88  3.48 
4.3 Asymmetric growth and hypertension
We investigated maternal hypertension, which has been previously found to be a determinant of the HCtoBW ratio in its severe, pregnancyinduced form (Kramer et al., 1990a). While information on hypertension is available in the VON data, unfortunately this variable has two limitations: it includes both chronic and pregnancyinduced hypertension (PIH), and is missing for about (though mostly in early years of data collection). Yet, some interesting observations can be made.
Hypertension is known to increase the likelihood of growth restriction. This is apparent from Figure 3 which shows estimated DQEs conditional on hypertension status. Its relationship with mortality risk is, however, controversial. Some studies suggested that PIH increases the risk of fetal, perinatal, and early neonatal mortality (Jain, 1997), while other studies found the opposite (Chen et al., 2006). In our data, the prevalence of hypertension (chronic and gestational) is overall about , while infant mortality rates are approximately and in, respectively, normotensive and hypertensive mothers, suggestive of a ‘protective’ effect of hypertension. However, the rate of hypertension is among mothers of babies with normal HCtoBW ratio, but in mothers of disproportionately small babies. In other words, there is a strong, positive association between hypertension and supranormal ratios ( test’s value ). As shown in Table 5, the mortality risk in infants with normal BW and HC is lower if born to hypertensive mothers as compared to normotensive mothers, regardless of their HCtoBW ratio. However, disproportionately small infants born to hypertensive mothers have an absolute risk of which, compared to an absolute risk of in their peers born to normotensive mothers, gives a rather different picture of the association between mortality and hypertension. It has been speculated that, in preterm infants born to hypertensive mothers, maternal hypertension is less damaging for fetal development than other causes of growth restriction (McBride et al., 2017). Our results do not exclude this hypothesis, but they also point to an interaction between hypertension (presumably its severe forms) and asymmetric growth restriction.
BW  HC  Risk  Lower  Upper  Risk  Lower  Upper  

Normal HCtoBW ratio  Normotensive  Hypertensive  
Normal  Normal  76,700  0.13  0.13  0.13  22,136  0.07  0.07  0.08 
10th  Normal  666  2.65  2.39  2.94  1,631  2.05  1.81  2.32 
10th  10th  1,972  2.43  2.28  2.60  3,479  3.30  3.06  3.56 
Supranormal HCtoBW ratio  Normotensive  Hypertensive  
Normal  Normal  3,993  0.17  0.16  0.19  3,353  0.07  0.06  0.08 
10th  Normal  951  1.80  1.61  2.03  2,080  2.75  2.36  3.22 
10th  10th  579  2.40  2.14  2.70  1,041  6.21  5.37  7.19 
5 Discussion
In this study, we proposed an approach to risk classification of abnormal ratios based on the allometric direction, which is intimately connected to directional quantiles (Kong and Mizera, 2012), and we applied these methods to data on birthweight and head circumference in a large cohort of preterm infants. Our analysis suggests that small preterm infants with large HCtoBW ratio are at increased mortality risk as compared not only to AGA infants, but also to proportionately growthrestricted preterm infants. There is evidence in the literature that asymmetric growth restriction increases the likelihood of adverse outcomes. One study concluded that “the prognosis of SGA infants with asymmetric growth [defined by the ratio of HC to abdominal circumference] is poorer than that of symmetrically grown infants and much worse than that of AGA infants” (Dashe et al., 2000). Therefore, our findings are consistent with the literature.
Following our approach, disproportionately growthrestricted infants can be identified as those whose BW and HC lie outside a given quantile contour along the allometric direction. It is straightforward to carry out the identification task. All is needed for classification is the scaling exponent and the cutoff . The calculation of the DQE is not necessary. Suppose for instance that an infant is born at 23 weeks and that her BW is 390 g and HC is 19.5 cm. These measurements already put her in a highrisk category since both measurements are below their respective 10th percentile (univariate) thresholds for gestational age and sex (Boghossian et al., 2016, 2018b). Using gestationalage specific estimates from Table 3, the allometric HCtoBW ratio is easily found as which is greater than the cutoff . The statistical justification of this procedure comes from Proposition 1, which asserts that the allometric direction is tangent to the DQE and thus guarantees that all the points such that are a subset of (jointly abnormal). Our study offers not only a statistically principled approach to risk classification, but also largesample estimated cutoffs that can be immediately used by practitioners together with previously published anthropometric charts for BW and HC using the same data (Boghossian et al., 2016, 2018b).
There are some limitations in our data. Firstly, gestational age is subject to measurement error and the accuracy depends on the method of estimation as well as when the measurement is made. We believe that our proposed approach can be extended to account for this error when estimating directional quantiles and allometric directions. Another possible limitation in our analysis is the omission of covariates that might explain different allometric relationships. We partly made up for this deficiency by investigating hypertension, and found that the mortality risk is associated with an interaction between asymmetric growth restriction and hypertension. In a separate analysis (results not shown), we investigated the allometric model (8) conditional on ethnicity but no meaningful differences were found. Unfortunately, other potentially relevant covariates such as maternal age or parental weight and height, which are known to be associated with birthweight (Griffiths et al., 2007; Geraci, 2016), are not available in the VON data. There are, however, a number of inhospital morbidity outcomes which we will explore in a separate study. Furthermore, it may be relevant to extend the analysis to infants with subnormal HCtoBW ratios. Suboptimal head size at birth is known to be a risk factor for poor neurodevelopmental outcomes if it persists after birth (Hack et al., 1991; Kuban et al., 2009). This type of investigation would require followup information which is not available to us at this time.
The proposed approach can be extended to applications where models equivalent to (8) are expected to hold and the interest lies in the classification of ratios. These obviously include applications with other anthropometric ratios, such as the waisttohip ratio, which has been proposed as a predictor of newborn size (Brown et al., 1996), the waisttoheight ratio, which has been found associated with prehypertension (Djeric et al., 2017), and the subscapulartotriceps skinfolds, fat–masstoweight, and fat–masstofat–free mass ratio indices which are used as risk factors in stunted populations (Judd, RamirezZea and Stein, 2008). As we move out of anthropometry, we find several other potential applications, especially in the biomedical sciences. For example, insulin and Cpeptide are strongly correlated, and their ratio is a biomarker used to discriminate between insulinoma, a tumor of the pancreas, and injection of excessive insulin (surreptitious or inadvertent) (Lebowitz and Blumenthal, 1993). The ejection fraction is a measure of the ratio between the blood pumped in and out of the left ventricle. A low input volume indicates atrial insufficiency, a low output volume ventricular insufficiency, and a low ratio might be associated to heart failure in several respects (e.g., decline of the contractile function). The list goes on with the albumintoglobulin ratio (liver insufficiency, immunodeficiency, autoimmunity, infection, cancer); calciumtoalbumin ratio (malnutrition); albumintocreatinine ratio (kidney disease); cortisoltocortisone ratio (cardiovascular risk); calcitrioltocalcifediol ratio (renal efficiency) (Rotondi et al., 2018); oxigen extration ratio (haemodialysis efficacy); metabolite ratios (type 2 diabetes) (Molnos et al., 2018); and stable isotope ratios (diet) (O’Brien, 2015).
Acknowledgements
Marco Geraci was funded by an ASPIRE grant from the Office of the Vice President for Research at the University of South Carolina and by the National Institutes of Health – National Institute of Child Health and Human Development (Grant Number: 1R03HD08480701A1). The authors wish to thank: Linglong Kong for providing the R code to perform quantile estimation with multivariate data as described in Kong and Mizera (2012); Erika Edwards for her support and participation in discussions of earlier drafts of the manuscript; two anonymous reviewers for their valuable comments on an earlier draft of the manuscript. The University of Vermont’s committee on human research determined that this study using VON’s deidentified research repository was not human subjects research.
References
 Azzalini and DallaValle (1996) [author] Azzalini, AA. DallaValle, AA. (1996). The multivariate skewnormal distribution. Biometrika 83 715726.
 Bernstein et al. (2000) [author] Bernstein, IMI., Horbar, JDJ., Badger, GJG., Ohlsson, AA. Golan, AA. (2000). Morbidity and mortality among verylowbirthweight neonates with intrauterine growth restriction. American journal of Obstetrics and Gynecology 182 198206.
 BoccaTjeertes et al. (2014) [author] BoccaTjeertes, II., Bos, AA., Kerstjens, JJ., de Winter, AA. Reijneveld, SS. (2014). Symmetrical and asymmetrical growth restriction in pretermborn children. Pediatrics 133 e650e656.
 Boghossian et al. (2016) [author] Boghossian, NSN., Geraci, MM., Edwards, EME., Morrow, KAK. Horbar, JDJ. (2016). Anthropometric charts for infants born between 22 and 29 weeks’ gestation. Pediatrics. e20161641.
 Boghossian et al. (2018a) [author] Boghossian, NSN., Geraci, MM., Edwards, EME. Horbar, JDJ. (2018a). Neonatal and fetal growth charts to identify preterm infants ¡30 weeks gestation at risk of adverse outcomes. American Journal of Obstetrics and Gynecology 219 195.e1195.e14.
 Boghossian et al. (2018b) [author] Boghossian, NSN., Geraci, MM., Edwards, EME. Horbar, JDJ. (2018b). Morbidity and mortality in small for gestational age infants at 22 to 29 weeks’ gestation. Pediatrics. doi:10.1542/peds.20172533.
 Brown et al. (1996) [author] Brown, JEJ., Potter, JDJ., Jacobs, DRD., Kopher, RAR., Rourke, MJM., Barosso, GMG., Hannan, PJP. Schmid, LAL. (1996). Maternal waisttohip ratio as a predictor of newborn size: Results of the Diana project. Epidemiology 7 6266.
 Chen et al. (2006) [author] Chen, XKX., Wen, SWS., Smith, GG., Yang, QQ. Walker, MM. (2006). General obstetrics: Pregnancyinduced hypertension is associated with lower infant mortality in preterm singletons. BJOG: An International Journal of Obstetrics & Gynaecology 113 544551.
 Corruccini (1983) [author] Corruccini, RSR. (1983). Principal components for allometric analysis. American Journal of Physical Anthropology 60 451453.
 Dashe et al. (2000) [author] Dashe, JSJ., McIntire, DDD., Lucas, MJM. Leveno, KJK. (2000). Effects of symmetric and asymmetric fetal growth on pregnancy outcomes. Obstetrics & Gynecology 96 321327.
 Djeric et al. (2017) [author] Djeric, MirjanaM., Ilincic, BranislavaB., Cabarkapa, VeliborV., Radosavkic, IsidoraI., Trifu, AleksandraA. Todorovic, MasaM. (2017). Prehypertension, waisttoheight ratio and markers of kidney function in apparently healthy men. Atherosclerosis 263 e271.
 Geraci (2016) [author] Geraci, MM. (2016). Estimation of regression quantiles in complex surveys with data missing at random: An application to birthweight determinants. Statistical Methods in Medical Research 25 13931421.
 Geraci, Alston and Birch (2013) [author] Geraci, MM., Alston, RDR. Birch, JMJ. (2013). Median percent change: A robust alternative for assessing temporal trends. Cancer Epidemiology 37 843849.
 Geraci and Jones (2015) [author] Geraci, MM. Jones, MCM. (2015). Improved transformationbased quantile regression. Canadian Journal of Statistics 43 118132.
 Gilchrist (2000) [author] Gilchrist, WW. (2000). Statistical Modelling with Quantile Functions. Chapman & Hall/CRC, Boca Raton, FL.

Goodman and Kotz (1973)
[author]
Goodman, IRI. Kotz, SS.
(1973).
Multivariate
generalized normal distributions. Journal of Multivariate Analysis 3 204219.
 Griffiths et al. (2007) [author] Griffiths, L. J.L. J., Dezateux, C.C., Cole, T. J.T. J. the Millennium Cohort Study Child Health Group (2007). Differential parental weight and height contributions to offspring birthweight and weight gain in infancy. International Journal of Epidemiology 36 104107.
 Diverse Populations Collaborative Group (2005) [author] Diverse Populations Collaborative Group (2005). Weightheight relationships and body mass index: Some observations from the diverse populations collaboration. American Journal of Physical Anthropology 128 220229.
 Guellec et al. (2015) [author] Guellec, II., Marret, SS., Baud, OO., Cambonie, GG., Lapillonne, AA., Roze, JCJ.C., Fresson, JJ., Flamant, CC., Charkaluk, MLM.L., Arnaud, CC. Ancel, PYP.Y. (2015). Intrauterine growth restriction, head size at birth, and outcome in very preterm infants. The Journal of Pediatrics 167 975981.e2.
 Hack et al. (1991) [author] Hack, MM., Breslau, NN., Weissman, BB., Aram, DorothyD., Klein, NancyN. Borawski, ElaineE. (1991). Effect of very low birth weight and subnormal head size on cognitive abilities at school age. New England Journal of Medicine 325 231237.
 Heymsfield et al. (2007) [author] Heymsfield, Steven B.S. B., Gallagher, DympnaD., Mayer, LaurelL., Beetsch, JoelJ. Pietrobelli, AngeloA. (2007). Scaling of human body composition to stature: New insights into body mass index. The American Journal of Clinical Nutrition 86 8291.
 Horbar et al. (2012) [author] Horbar, JDJ., Carpenter, JHJ., Badger, GJG., Kenny, MJM., Soll, RFR., Morrow, KAK. Buzas, JSJ. (2012). Mortality and neonatal morbidity among infants 501 to 1500 grams from 2000 to 2009. Pediatrics 129 10191026.
 Jain (1997) [author] Jain, LL. (1997). Effect of pregnancyinduced and chronic hypertension on pregnancy outcome. Journal of Perinatology 17 425—427.
 Jolicoeur (1963) [author] Jolicoeur, PP. (1963). 193. Note: The multivariate generalization of the allometry equation. Biometrics 19 497499.
 Judd, RamirezZea and Stein (2008) [author] Judd, SES., RamirezZea, MM. Stein, ADA. (2008). Relation of ratio indices of anthropometric measures to obesity in a stunted population. American Journal of Human Biology 20 446450.
 Koenker and Bassett (1978) [author] Koenker, RR. Bassett, GG. (1978). Regression quantiles. Econometrica 46 3350.
 Kong and Mizera (2012) [author] Kong, LL. Mizera, II. (2012). Quantile tomography: Using quantiles with multivariate data. Statistica Sinica 22 15891610.
 Kozubowski and Podgorski (2000) [author] Kozubowski, TJT. Podgorski, KK. (2000). A multivariate and asymmetric generalization of Laplace distribution. Computational Statistics 15 531540.
 Kozubowski, Podgórski and Rychlik (2013) [author] Kozubowski, TJT., Podgórski, KK. Rychlik, II. (2013). Multivariate generalized Laplace distribution and related random fields. Journal of Multivariate Analysis 113 5972.
 Kramer et al. (1990a) [author] Kramer, MSM., Olivier, MM., McLean, FHF., Dougherty, GEG., Willis, DMD. Usher, RHR. (1990a). Determinants of fetal growth and body proportionality. Pediatrics 86 1826.
 Kramer et al. (1990b) [author] Kramer, MSM., Olivier, MM., McLean, FHF., Willis, DMD. Usher, RHR. (1990b). Impact of intrauterine growth retardation and body proportionality on fetal and neonatal outcome. Pediatrics 86 707713.
 Kuban et al. (2009) [author] Kuban, KCKK., Allred, ENE., O’Shea, TMT., Paneth, NN., Westra, SS., Miller, CC., Rosman, NPN. Leviton, AA. (2009). Developmental correlates of head circumference at birth and two years in a cohort of extremely low gestational age newborns. The Journal of Pediatrics 155 344349.e3.
 Lebowitz and Blumenthal (1993) [author] Lebowitz, MRM. Blumenthal, SAS. (1993). The molar ratio of insulin to Cpeptide: An aid to the diagnosis of hypoglycemia due to surreptitious (or inadvertent) insulin administration. Archives of Internal Medicine 153 650655.
 Lin, Su and River (1991) [author] Lin, CCC.C., Su, SJS.J. River, LPL. (1991). Comparison of associated highrisk factors and perinatal outcome between symmetric and asymmetric fetal intrauterine growth retardation. American Journal of Obstetrics & Gynecology 164 15351542.
 Mao, Wei and Liu (2017) [author] Mao, GG., Wei, YY. Liu, YY. (2017). SIMEX method for censored quantile regression with measurement error. Communications in Statistics  Simulation and Computation 46 75527560.
 McBride et al. (2017) [author] McBride, CAC., Bernstein, IMI., Badger, GJG. Soll, RFR. (2017). Maternal hypertension and mortality in small for gestational age 22 to 29week infants. Reproductive Sciences 25 276280.
 McKeague et al. (2011) [author] McKeague, IWI., LópezPintado, SS., Hallin, MM. Šiman, MM. (2011). Analyzing growth trajectories. Journal of Developmental Origins of Health and Disease 2 322329.
 Molnos et al. (2018) [author] Molnos, SS., Wahl, SS., Haid, MM., Eekhoff, EMWE., Pool, RR., Floegel, AnnaA., Deelen, JorisJ., Much, DanielaD., Prehn, CorneliaC., Breier, MichaelaM., Draisma, Harmen H.H. H., van Leeuwen, NienkeN., SimonisBik, Annemarie M. C.A. M. C., Jonsson, AnnaA., Willemsen, GonnekeG., Bernigau, WolfgangW., WangSattler, RuiR., Suhre, KarstenK., Peters, AnnetteA., Thorand, BarbaraB., Herder, ChristianC., Rathmann, WolfgangW., Roden, MichaelM., Gieger, ChristianC., Kramer, Mark H. H.M. H. H., van Heemst, DianaD., Pedersen, Helle K.H. K., Gudmundsdottir, ValborgV., Schulze, Matthias B.M. B., Pischon, TobiasT., de Geus, Eco J. C.E. J. C., Boeing, HeinerH., Boomsma, Dorret I.D. I., Ziegler, Anette G.A. G., Slagboom, P. ElineP. E., Hummel, SandraS., Beekman, MarianM., Grallert, HaraldH., Brunak, SørenS., McCarthy, Mark I.M. I., Gupta, RamneekR., Pearson, Ewan R.E. R., Adamski, JerzyJ. ’t Hart, Leen M.L. M. (2018). Metabolite ratios as potential biomarkers for type 2 diabetes: a DIRECT study. Diabetologia 61 117129.
 O’Brien (2015) [author] O’Brien, DMD. (2015). Stable isotope ratios as biomarkers of diet for health research. Annual Review of Nutrition 35 565594.
 Olsen et al. (2009) [author] Olsen, IEI., Lawson, MLM., MeinzenDerr, JJ., Sapsford, ALA., Schibler, KRK., Donovan, EFE. Morrow, ALA. (2009). Use of a body proportionality index for growth assessment of preterm infants. The Journal of Pediatrics 154 486491.
 Quetelet (1842) [author] Quetelet, MAM. (1842). A treatise on man and the development of his faculties. W. and R. Chambers, Edinburgh.
 Rotondi et al. (2018) [author] Rotondi, SS., Tartaglione, LL., Muci, MLM., Farcomeni, AA., Pasquali, MM. Mazzaferro, SS. (2018). Oxygen extraction ratio (OER) as a measurement of hemodialysis (HD) induced tissue hypoxia: A pilot study. Scientific Reports 8 5655.
 Saleem et al. (2011) [author] Saleem, TT., Sajjad, NN., Fatima, SS., Habib, NN., Ali, SRS. Qadir, MM. (2011). Intrauterine growth retardation–small events, big consequences. Italian Journal of Pediatrics 37 41.
 Somers (1986) [author] Somers, KMK. (1986). Multivariate allometry and removal of size with principal components analysis. Systematic Zoology 35 359368.
 Stoll et al. (2010) [author] Stoll, BJB., Hansen, NIN., Bell, EFE., Shankaran, SS., Laptook, ARA., Walsh, MCM., Hale, ECE., Newman, NSN., Schibler, KK. Carlo, WAW. (2010). Neonatal outcomes of extremely preterm infants from the NICHD Neonatal Research Network. Pediatrics 126 443456.
 Vandenbosche and Kirchner (1998) [author] Vandenbosche, RCR. Kirchner, JTJ. (1998). Intrauterine growth retardation. American Family Physician 58 138490, 13934.
 Warton et al. (2006) [author] Warton, DID., Wright, IJI., Falster, DSD. Westoby, MM. (2006). Bivariate linefitting methods for allometry. Biological Reviews 81 259291.
 Warton et al. (2012) [author] Warton, DID., Duursma, RAR., Falster, DSD. Taskinen, SS. (2012). smatr 3 – an R package for estimation and inference about allometric lines. Methods in Ecology and Evolution 3 257259.
 Wei (2008) [author] Wei, YY. (2008). An approach to multivariate covariatedependent quantile contours with application to bivariate conditional growth charts. Journal of the American Statistical Association 103 397409.
 Wei and Carroll (2009) [author] Wei, YY. Carroll, RJR. (2009). Quantile regression with measurement error. Journal of the American Statistical Association 104 11291143.
 Williams and O’Brien (1998) [author] Williams, MCM. O’Brien, WFW. (1998). Low weight/length ratio to assess risk of cerebral palsy and perinatal mortality in twins. American Journal of Perinatology 11 225228.
Appendix A Appendix – Multivariate percentiles and allometric analysis
Let us consider a dimensional vector , , and the elementwise logtransformed vector . Also, let be the positivedefinite variancecovariance matrix of . The goal of PCA applied to is to obtain the decomposition . (As noted elsewhere (Jolicoeur, 1963), the logtransformation removes measurement scale differences and therefore makes the use of the correlation matrix unnecessary. However, some authors (Somers, 1986) advocate the use of the correlation matrix to separate size and shape variation.) The columns of , say , , are the unitnorm eigenvectors of the decomposition, while the diagonal matrix has diagonal elements ,
, and are given by the corresponding eigenvalues. The PCA scores are then obtained as
.The first component (major axis) is the direction with maximal variance and its equation can expressed as (Jolicoeur, 1963)
where is the angle made by the first principal component with the coordinate axis of . Thus
where
is the geometric mean of
. It follows that(11) 
where and . That is, the ratio of any pair of PCA loadings, say and , from the first component approximates the slope of the bivariate MA regression for the variables and (Corruccini, 1983). Using these results, we offer the following proposition.
Proposition 2.
Let be a multivariate random vector collecting continuous and strictly positive random variables, and assume that the PCA allometric model (11) holds true for any pair of variables , with . Also, define and assume that the th directional quantile envelope generated by
is smooth. Then the hyperplanes
(12) 
where and , are tangent to .
Proof.
In its implicit form, the hyperplane tangent to is by definition given by
for any given direction . We also note that
(13) 
On the logscale, the allometric equations given in Proposition 2 relating a given to the other variables , , can be rewritten as
If we take the sum of the terms on the left and righthand sides, respectively, we obtain
(14) 
which has the same form of (13) with , for all , and . Therefore, the hyperplane tangent to in the allometric direction, defined as , is . By (13) we have that
Similarly to before, we find . Then the tangent hyperplane becomes , which corresponds to the first equation given in (2). The second equation in (2) is found with similar arguments as those used in Proposition 1. ∎
Model (11) is rather flexible as it allows for distinct slopes to be estimated. If there is evidence or theoretical justification of equality of slopes, some restrictions can be imposed, for example, by assuming homogeneity of the ’s for some ’s or using an even more parsimonious model of the kind , , where indexes a subset of variables , .
Appendix B Appendix – Additional tables
BW  HC  Risk  Lower  Upper  

weeks  
Normal  Normal  4,655  0.52  0.51  0.54 
Normal  10th  322  1.23  1.13  1.34 
10th  Normal  306  1.36  1.26  1.47 
10th  10th  213  1.50  1.39  1.61 
weeks  
Normal  Normal  14,979  0.19  0.19  0.20 
Normal  10th  650  1.45  1.27  1.65 
10th  Normal  736  1.85  1.67  2.05 
10th  10th  1,033  2.69  2.51  2.88 
weeks  
Normal  Normal  19,571  0.06  0.06  0.07 
Normal  10th  649  1.84  1.47  2.30 
10th  Normal  860  2.12  1.77  2.54 
10th  10th  1,368  4.19  3.77  4.65 
weeks  
Normal  Normal  24,318  0.02  0.02  0.03 
Normal  10th  995  1.15  0.79  1.69 
10th  Normal  1,253  1.70  1.28  2.25 
10th  10th  1,531  3.83  3.21  4.58 
BW  HC  Risk  Lower  Upper  

weeks  
Normal  Normal  5,161  0.59  0.58  0.60 
Normal  10th  297  1.18  1.10  1.28 
10th  Normal  351  1.29  1.21  1.37 
10th  10th  227  1.42  1.33  1.51 
weeks  
Normal  Normal  16,721  0.24  0.23  0.24 
Normal  10th  601  1.44  1.29  1.62 
10th  Normal  953  1.68  1.55  1.83 
10th  10th  1,030  2.36  2.22  2.50 
weeks  
Normal  Normal  21,623  0.08  0.08  0.09 
Normal  10th  934  1.47  1.23  1.76 
10th  Normal  954  2.08  1.79  2.40 
10th  10th  1,485  3.99  3.67  4.35 
weeks  
Normal  Normal  27,565  0.03  0.03  0.03 
Normal  10th  1,247  1.63  1.26  2.11 
10th  Normal  1,348  1.79  1.41  2.27 
10th  10th  1,810  4.52  3.94  5.18 
BW  HC  Risk  Lower  Upper  Risk  Lower  Upper  

Females  Normal HCtoBW ratio  Supranormal HCtoBW ratio  
Normal  Normal  61,254  0.11  0.11  0.11  2,269  0.20  0.18  0.21 
10th  Normal  405  0.93  0.70  1.25  2,750  0.91  0.61  1.37 
10th  10th  582  1.87  1.59  2.20  3,563  1.11  0.99  1.23 
Males  Normal HCtoBW ratio  Supranormal HCtoBW ratio  
Normal  Normal  70,223  0.13  0.13  0.14  847  0.26  0.23  0.29 
10th  Normal  1,055  1.56  1.38  1.75  2,551  1.03  0.48  2.21 
10th  10th  1,098  1.86  1.68  2.07  3,454  1.02  0.89  1.16 
Comments
There are no comments yet.