Skewed distributions, determination

The major urinary metabolite of di(2-ethylhexyl) adipate, 2-ethylhexanoic acid, has been shown to be an appropriate marker for biological monitoring of dietary di(2-ethylhexyl) adipate intake (Loftus etal., 1993, 1994). A limited population study in the United Kingdom was undertaken to estimate the daily intake of di(2-ethylhexyl) adipate following intake of a mean dose of 5.4 mg di(2-ethylhexyl) adipate presented with food. The study involved the determination of the urinary metabolite, 2-ethyl-hexanoic acid (24-h mine sample) in 112 individuals from five geographical locations. A skewed distribution with a median value for the daily intake of 2.7 mg was determined (Loftus et al., 1994). This value is about one third of the indirectly estimated maximum intake of 8. 2 mg per day. The probability of a daily intake in excess of 8.2 mg in the limited population (112 individuals) was calculated to be 3% (Loftus etal, 1994). 

In 1991 the UK Committee on Toxicology (COT) set a tolerable daily intake value (TDI) of 0.3 mg/kg body weight/day for DEHA. One year later, in 1992, a urinary biomarker study was reported for DEHA in a limited population exercise in the UK. A skewed distribution was determined with a median value of 2.7 mg/day and this confirmed by an independent route, the earlier estimates of DEHA intake made using dietary survey data. 

Colloidal systems are generally of a polydispersed nature - i.e. the molecules or particles in a particular sample vary in size. By virtue of their stepwise build-up, colloidal particle and polymer molecular sizes tend to have skew distributions, as illustrated in Figure 1.2, for which the Poisson distribution often offers a good approximation. Very often, detailed determination of relative molecular mass or particle size distribution is impracticable and less perfect experimental methods, which yield average values, must be accepted. The significance of the word average depends on the relative contributions of the various molecules or particles to the property of the system which is being measured. 

Precision determines the reproducibility or repeatability of the analytical data. It measures how closely multiple analysis of a given sample agree with each other. If a sample is repeatedly analyzed under identical conditions, the results of each measurement, x, may vary from each other due to experimental error or causes beyond control. These results will be distributed randomly about a mean value which is the arithmetic average of all measurements. If the frequency is plotted against the results of each measurement, a bell-shaped curve known as normal distribution curve or gaussian curve, as shown below, will be obtained (Figure 1.2.1). (In many highly dirty environmental samples, the results of multiple analysis may show skewed distribution and not normal distribution.)... 

Maxent is one method for determining the probability of a model. A simple example is that of the toss of a six sided unbiassed die. What is the most likely underlying frequency distribution, and how can each possible distribution be measured Figure 3.31 illustrates a flat distribution and Figure 3.32 a skew distribution (expressed... 

The mode and the median may be determined graphically but the above summation has to be carried out for the determination of the mean. For a slightly skewed distribution the approximate relationship, meanmode = 3(mean-median) holds. For a symmetrical distribution, all three averages coincide. These means represent the distribution in only two of its properties. The characteristics of a particle size distribution are its total number, length, surface, volume (mass) and moment. Note that ... 

Although many of the components of a cost estimate are skewed distributions, when these are combined the resulting distribution is often approximately normal. The preceding guidelines can thus be used to determine the amount of contingency charge needed for a given level of confidence. 

Median The median is the middle of a distribution half the scores are above the median and half are below it. Unlike the mean, the median is not highly sensitive to extreme data points. This makes the median a better measure than the mean for finding the central tendency of highly skewed distributions. The median is determined by organizing the data points from lowest to highest. When there is an odd number of numbers, the median is simply the middle number. For example, the median of 2, 4, and 7 is 4. When there is an even number of numbers, the median is the mean of the two middle numbers. Thus, the median of the numbers 2, 4, 7, 12 is (4+7)/2 = 5.5. 

Checking the skewness distribution for outliers provides a fast method to determine if few individual structures do not fit in the data set (Fignre 6.13). The three ontliers with a descriptor skewness of 4.0 and higher are (1) hydrazine (2) thionyl chloride and (3) ammonium chloride — three compounds that are not representative for the wide variation of organic strnctnres in the remaining data set. 

Based on the estimated density functions, the mean and the 95%-quantile are superimposed by the dashed lines in Figure 4-13a and Figure 4.13b whereby the latter coincides with the re-order level ensuring a 95% a-service level. For both sites the density functions of total consumption are bimodal and skewed to the right. The first mode is at zero consumption and is inherited from the Markov chain of the production models. It corresponds to situations when a pipeline inspection coincides with a cracker. shut-down. The second mode is inherited from the Weibull distribution determining the pipeline inspection time which also causes the skewness. 

In the density function, a determines the shape, P determines the skewness, p determines the location, and A, determines the heaviness of the tails. 5 is the scaling parameter, which is comparable to a in the normal distribution. 

The skew, denoted by y, measures the amount of asymmetry in the distribution. Skewness is determined by examining the relationship in the clustering of extreme values, that is, the tails. If more of the data set is clustered towards the smaller extreme values, then it is said that the system has positive or right skewness. On the other hand, if the data set is clustered towards the larger extreme values, then it is said that the system has negative or left skewness. The skew of a data set can be computed as... 

Replot the laboratory and field data using a dimensionless scale for the abscissa based on the ratio of the variable initiating parameter X to the characteristic zero-incidence value Xq, e.g., the curves 04B and OCDE as shown in Fig. 7, and determine the constants of a suitable skewed distribution equation, such as the modified first derivative of Eq. (14), with exponent n equal to 2 ... 

The median is defined as the diameter for which one-half the total munber of particles are smaller and one-half are larger. The median is also the diameter that divides the frequency distribution curve into equal areas, and the diameter corresponding to a cumulative fraction of 0.5. The mode is the most frequent size, or the diameter associated with the highest point on the frequency function curve. The mode can be determined by. setting the derivative of the frequency function equal to zero and solving for d. For symmetrical distributions such as the normal distribution, the mean, median, and mode will have the same value, which is the diameter of the axis of synunetry. For an asymmetrical or skewed distribution, these quantities will have different values. The median is conunonly used with skewed distributions, because extreme values in the tail have less effect on the median than on the mean. Most aerosol size distributions are skewed, with a long tail to the right, as shown in Fig. 4.4. For such a distribution,... 

Thus the two parameters of the distribution can be determined after finding the variance and skewness from the experimental RTD. 

The distribution of Lp(a) concentrations in white populations is highly skewed, but does not differ significantly between males and females (Fig. 7). The median Lp(a) value determined in a Belgian population sample is 0.140 g/liter (L3), in agreement with values reported in other European and Caucasian American populations (H26). 

Currently, the EPA considers acute pesticide exposures to represent a reasonable certainty of no harm when exposure at the 99.9th percentile is below the RfD. When exposure at the 99.9th percentile exceeds the RfD, the EPA will generally conduct a sensitivity analysis to determine whether particular factors that drive the exposure, such as high residue or high consumption levels, are unusual and so may represent artifacts that artificially skew the exposure distribution curve. 

---