Multiple diagnostic statistics

Multiple overlapping, prospective methods of case ascertainment used to identify all individuals with condition of interest from a predefined population includes searches of both primary and secondary care and databases of diagnostic tests and death certification/mortality statistics... 

The basic principle of experimental design is to vary all factors concomitantly according to a randomised and balanced design, and to evaluate the results by multivariate analysis techniques, such as multiple linear regression or partial least squares. It is essential to check by diagnostic methods that the applied statistical model appropriately describes the experimental data. Unacceptably poor fit indicates experimental errors or that another model should be applied. If a more complicated model is needed, it is often necessary to add further experimental runs to correctly resolve such a model. 

Notes It is recommended that statistical and computer analysis of kinetic data should be carried out to evaluate kinetic parameters (Cleland, 1967 Cornish-Bowden, 1995), however these linear plots may still retain their diagnostic values. The linear plots indicate the compliance to die Michaelis-Menten kinetics whereas nonlinear plots imply multiple substrate addition, substrate inhibition or homotropic allosterism. 

Finally some comments are called for on judging the comparison of fitting different equations to the same experimental data. With some notable exceptions, such as the dissection of multiple spectral changes, the use of elaborate statistical diagnostics raises the suspicion of inadequate experimentation. However, the simple procedure of comparing the residuals when different equations - say the sums of one, two, or three exponentials - are used, is recommended. The eye is the best judge of systematic deviations along the time axis ... 

Monte Carlo Simulation Under most circumstances, models that are implied by data are non-unique. Many different models integrate the same data. Hence a probabilistic description is appropriate. This is reviewed in detail later. However, it requires the results from multiple simulations to assess the implied statistics of the diagnostic functions. 

---