Changes in measurement units and baseline correspond to affine transformations on . We write
The fact that the mean of (9.7) cannot be defined for all distributions is an indication of its lack of robustness. More precisely the functional is not locally bounded (9.11) in the metric at any distribution . The median MED can be defined at any distribution as the mid-point of the interval of -values for which
An important family of statistical functionals is the family of M-functionals introduced by [56] Let and be functions defined on with values in the interval . For a given probability distribution we consider the following two equations for and
In order to guarantee existence and uniqueness conditions have to be placed on the functions and as well as on the probability measure . The ones we use are due to [99] (see also [58]) and are as follows:
( ) | for all . |
( ) | is strictly increasing |
( ) | |
( ) | is continuously differentiable with derivative . |
( ) | for all . |
( ) | is strictly increasing |
( ) | |
( ) | |
( ) | is continuously differentiable with derivative . |
( ) | is strictly increasing. |
If these conditions hold and satisfies
The main disadvantage of M-functionals defined by (9.21) and (9.22) is ( ) which links the location and scale parts in a manner which may not be desirable. In particular there is a conflict between the breakdown behaviour and the efficiency of the M-functional (see below). There are several ways of overcoming this. One is to take the scale function and then to calculate a second location functional by solving
In some situations there is an interest in downweighting outlying observations completely rather than in just bounding their effect. A downweighting to zero is not possible for a -function which satisfies ( ) but can be achieved by using so called redescending -functions such as Tukey's biweight
So far all scale functionals have been defined in terms of a deviation from a location functional. This link can be broken as follows. Consider the functional defined to be the solution of
Although we have defined -functionals as a solution of (9.21) and (9.22) there are sometimes advantages in defining them as a solution of a minimization problem. Consider the Cauchy distribution with density
Another class of functionals defined by a minimization problem is the class of -functionals. Given a function which is symmetric, continuous on the right and non-increasing on with and . We define by
Given a location functional the bias is defined by
The breakdown point of at with respect to is defined by
For location and scale functionals there exist upper bounds for
the breakdown points. For location functionals we have
We refer to [58]. It may be shown that all breakdown points of the mean are zero whereas the median attains the highest possible breakdown point in each case.The corresponding result for scale functionals is more complicated. Whereas we know of no reasonable metric in (9.42) of Theorem 1 which leads to a different upper bound this is not the case for scale functionals. [58] shows that for the Kolmogoroff metric the corresponding upper bound is but is for the gross error neighbourhood. If we replace the Kolmogoroff metric by the standard Kuiper metric defined by
Similarly all breakdown points of the standard deviation are zero but, in contrast to the median, the MAD does not attain the upper bounds of (9.44). We have
The M-functional defined by (9.21) and (9.22) has a breakdown point which satisfies
The breakdown point is a simple but often effective measure of the robustness of a statistical functional. It does not however take into account the size of the bias. This can be done by trying to quantify the minimum bias over some neighbourhood of the distribution and if possible to identify a functional which attains it. We formulate this for and consider the Kolmogoroff ball of radius . We have ([58])
In other words the median minimizes the bias over any Kolmogoroff neighbourhood of the normal distribution. This theorem can be extended to other symmetric distributions and to other situations (Riedel, 1989a, 1989b). It is more difficult to obtain such a theorem for scale functionals because of the lack of a property equivalent to symmetry for location. Nevertheless some results in this direction have been obtained and indicate that the length of the shortest half of (9.8) has very good bias properties ([74]).
Given a sample with empirical measure we can calculate a location functional which in some sense describes the location of the sample. Such a point value is rarely sufficient and in general should be supplemented by a confidence interval, that is a range of values consistent with the data. If is differentiable (9.12) and the data are i.i.d. random variables with distribution then it follows from (9.3) (see Sect. 9.1.3) that an asymptotic -confidence interval for is given by
The precision of the functional at the distribution can be quantified by the length of the asymptotic confidence interval (9.51). As the only quantity which depends on is we see that an increase in precision is equivalent to reducing the size of . The question which naturally arises is then that of determining how small can be made. A statistical functional which attains this lower bound is asymptotically optimal and if we denote this lower bound by , the efficiency of the functional can be defined as . The efficiency depends on and we must now decide which or indeed s to choose. The arguments given in Sect. 9.1.2 suggest choosing a which maximizes over a class of models. This holds for the normal distribution which maximizes over the class of all distributions with a given variance. For this reason and for simplicity and familiarity we shall take the normal distribution as the reference distribution. If a reference distribution is required which also produces outliers then the slash distribution is to be preferred to the Cauchy distribution. We refer to [19] and the discussion given there.
If we consider the M-functionals defined by (9.24) and (9.25) the efficiency at the normal distribution is an increasing function of the tuning parameter . As the breakdown point is a decreasing function of this would seem to indicate that there is a conflict between efficiency and breakdown point. This is the case for the M-functional defined by (9.24) and (9.25) and is due to the linking of the location and scale parts of the functional. If this is severed by, for example, recalculating a location functional as in (9.26) then there is no longer a conflict between efficiency and breakdown. As however the efficiency of the location functional increases the more it behaves like the mean with a corresponding increase in the bias function of (9.35) and (9.37). The conflict between efficiency and bias is a real one and gives rise to an optimality criterion, namely that of minimizing the bias subject to a lower bound on the efficiency. We refer to [73].
One of the main uses of robust functionals is the labelling of so called outliers (see [5], [55], [3], [40], [42], and Simonoff (1984, 1987)). In the data of Table 9.1 the laboratories 1 and 3 are clearly outliers which should be flagged. The discussion in Sect. 9.1.1 already indicates that the mean and standard deviation are not appropriate tools for the identification of outliers as they themselves are so strongly influenced by the very outliers they are intended to identify. We now demonstrate this more precisely. One simple rule is to classify all observations more than three standard deviations from the mean as outliers. A simple calculation shows that this rule will fail to identify arbitrarily large outliers with the same sign. More generally if all observations more than standard deviations from the mean are classified as outliers then this rule will fail to identify a proportion of outliers with the same sign. This is known as the masking effect ([79]) where the outliers mask their presence by distorting the mean and, more importantly, the standard deviation to such an extent as to render them useless for the detection of the outliers. One possibility is to choose a small value of but clearly if is too small then some non-outliers will be declared as outliers. In many cases the main body of the data can be well approximated by a normal distribution so we now investigate the choice of for samples of i.i.d. normal random variables. One possibility is to choose dependent on the sample size so that with probability say 0.95 no observation will be flagged as an outlier. This leads to a value of of about ([29]) and the largest proportion of one-sided outliers which can be detected is approximately which tends to zero with . It follows that there is no choice of which can detect say outliers and at the same time not falsely flag non-outliers. In order to achieve this the mean and standard deviation must be replaced by functionals which are less effected by the outliers. In particular these functionals should be locally bounded (9.11). Considerations of asymptotic normality or efficiency are of little relevance here. Two obvious candidates are the median and MAD and if we use them instead of the mean and standard deviation we are led to the identification rule ([53]) of the form
(9.61) |
We can now formulate the task of outlier identification for the normal distribution as follows: For a given sample which contains at least i.i.d. observations distributed according to , we have to find all those that are located in . The level can be chosen to be dependent on the sample size. If for some we set
(9.63) |
To describe the worst case behaviour of an outlier identifier we can look at the largest nonidentifiable outlier, which it allows. From [29] we report some values of this quantity for the Hampel identifier (HAMP) and contrast them with the corresponding values of a sophisticated high breakdown point outwards testing identifier (ROS), based on the non-robust mean and standard deviation ([87]; [107]). Both identifiers are standardized by (9.65) with . Outliers are then observations with absolute values greater than , and . For outliers and the average sizes of the largest non-detected outlier are 6.68 (HAMP) and 8.77 (ROS), for outliers and the corresponding values are 4.64 (HAMP) and 5.91 (ROS) and finally for outliers and the values are 5.07 (HAMP) and 9.29 (ROS).