.. _DiscImplausibilityCutoff: Discussion: Implausibility Cutoffs ================================== Description and Background -------------------------- As introduced in :ref:`ThreadGenericHistoryMatching`, an integral part of the :ref:`history matching` process is the imposing of cutoffs on the various :ref:`implausibility measures` introduced in :ref:`AltImplausibilityMeasure`. This page discusses the relevant considerations involved in choosing the cutoffs. The notation used is the same as that defined in :ref:`ThreadGenericHistoryMatching`. Here we use the term model synonymously with the term :ref:`simulator`. Discussion ---------- History matching attempts to identify the set :math:`\mathcal{X}` of all inputs that would give rise to acceptable matches between model outputs and observed data. This is achieved by imposing cutoffs on implausibility measures in order to discard inputs :math:`x` that are highly unlikely to belong to :math:`\mathcal{X}`. The specific values used for these cutoffs is therefore important, and several factors must be considered before choosing them. Univariate Cutoffs ~~~~~~~~~~~~~~~~~~ Consider the univariate implausibility measure :math:`I_{(i)}(x)` corresponding to the :math:`i`th output of a multi-output function :math:`f(x)`, introduced in :ref:`AltImplausibilityMeasure`: .. math:: I^2_{(i)}(x) = \frac{ ({\rm E}[f_i(x)] - z_i )^2}{ {\rm Var}[ {\rm E}[f_i(x)] -z_i] } = \frac{ ({\rm E}[f_i(x)] - z_i )^2}{{\rm Var}[f_i(x)] + {\rm Var}[d_i] + {\rm Var}[e_i]} where the second equality follows from the definition of the best input approach (see :ref:`DiscBestInput` for details). We are to impose a cutoff :math:`c` such that values of :math:`x` that satisfy: .. math:: I_{(i)}(x) \le c are to be analysed further, and all other values of :math:`x` are discarded. This defines a new sub-volume of the input space that we refer to as the non-implausible volume. Determining a reasonable value for the cutoff :math:`c` can be achieved using certain unimodality arguments, which are employed as follows. Regarding the size of the individual univariate implausibility measure :math:`I_{(i)}(x)`, we again consider :math:`x` as a candidate for the best input :math:`x^+`. If we then make the fairly weak assumption that the appropriate distribution of :math:`({\rm E}[f_i(x)]-z)`, with :math:`x=x^+ `, is both unimodal and continuous, then we can use the :math:`3\sigma` rule (Pukelsheim, 1994) which implies quite generally that :math:`I_{(i)}(x) \le 3` with probability greater than 0.95, even if the distribution is heavily asymmetric. A value of :math:`I_{(i)}(x)` greater than a cutoff of :math:`c=3` would suggest that the input :math:`x` could be discarded. Consideration of the fraction of input space that is removed for different choices of :math:`c` may alter this value. For example, if we find that we can remove a sufficiently large percentage of the input space with a more conservative choice of say :math:`c=4`, then we may adopt this value instead. In addition, we may also perform various diagnostics to check that we are not discarding any acceptable runs (Vernon, 2010). Multivariate Cutoffs ~~~~~~~~~~~~~~~~~~~~ We can use the above univariate cutoff :math:`c` as a guide to determine an equivalent cutoff :math:`c_M` for the maximum implausibility measure such that we discard all :math:`x` that do not satisfy .. math:: I_{M}(x) \le c_M If we are dealing with a small number of highly correlated outputs, then this will be similar to the univariate case and we can use Pukelsheim's 3 sigma rule and choose values for :math:`c_M` that are similar to or slightly larger than :math:`c=3` (e.g. :math:`c_M = 3.2, 3.5` say). If there are a large number of outputs, and we believe them to be mostly uncorrelated then a higher value must be chosen (e.g. :math:`c_M = 4`, 4.5 or 5 say) depending on the fraction of space cut out. If we want to be more precise, we can make further assumptions as to the shape of the individual distributions that are used in the implausibility measures: for example we can assume they are normally distributed, and look up suitable tables for the maximum of a collection of independent or even correlated normals at certain conservative significance levels (we would recommend 0.99 or higher). Similar considerations can be used to generate sensible cutoffs for the second and third maximum implausibility measures. See :ref:`DiscInformalAssessMD` for similar discussions regarding cutoffs, and Vernon, 2010 for examples of their use. Consider the full multivariate implausibility measure :math:`I_{MV}(x)` introduced in :ref:`AltImplausibilityMeasure`: .. math:: I_{MV}(x) = (z - {\rm E}[f(x)])^T ({\rm Var}[f(x)] + {\rm Var}[d] + {\rm Var}[e])^{-1} (z - {\rm E}[f(x)]) Choosing a suitable cutoff :math:`c_{MV}` for :math:`I_{MV}(x)` is more complicated. As a simple heuristic, we might choose to compare :math:`I_{MV}(x)` with the upper critical value of a :math:`\chi^2`-distribution with degrees-of-freedom equal to the number of outputs considered. This should then be combined with considerations of percentage space cutout along with diagnostics as discussed above. Additional Comments ------------------- In order to link the different cutoffs :math:`c`, :math:`c_M` and :math:`c_{MV}` rigorously, we would of course have to make full multivariate distributional assumptions over each of the relevant output quantities. This is a level of detail that is in many cases not necessary, provided relatively conservative choices for each of the cutoffs are made. The history matching process involves applying the above implausibility cutoffs iteratively, as is described in :ref:`ThreadGenericHistoryMatching` and discussed in detail in :ref:`DiscIterativeRefocussing`. The implausibility measures are simple and very fast to evaluate (as they only rely on evaluations of the emulator), and hence the application of the cutoffs is tractable even for testing large numbers of input points. References ---------- Pukelsheim, F. (1994). “The three sigma rule.” *The American Statistician*, 48: 88–91. Vernon, I., Goldstein, M., and Bower, R. (2010), “Galaxy Formation: a Bayesian Uncertainty Analysis,” MUCM Technical Report 10/03