Biomedical subjects
Publications and source records attributed to M E Tarter.
Explore the source record for details and available documents.
Five statistical procedures were used to partial the correlation between waterborne asbestos and digestive site cancer for the putative effects of population density. These include: analysis based on a data subset with roughly homogeneous population density; standard residual analysis (partial correlation); conditional probability integral transformation; analysis based upon ranked data, and use of logarithmic transformation. Nonparametric regression graphical techniques are applied to examine the nature or shape of the asbestos-cancer dose-response curve. Evidence is presented that suggests that there is considerable difference between analyses involving nonhigh-density tracts and non-San Francisco tracts. Evidence is also presented that the modal-type nonparametric regression curve forks or bifurcates when adjustment is made for population density.
Nonparametric estimates of joint, conditional and marginal probability densities can be used to estimate the relative probability of a data point's recurrence. Outlier, unusual or abnormal values of a random variate tend to be those which are unlikely to recur. As part of an interactive graphical system, a procedure has been implemented which enables a biomedical researcher to view both the estimated probability and the numerical value of a data point's coordinates. This display circumvents the problem of interpreting a normal range in two or more dimensions and can thus be more easily generalized than most alternative outlier detection procedures.
The role of "methodology", as distinguished from "theory" and "application", is discussed and illustrated. It is argued that research in the biomedical sciences is moving towards a degree of complexity different in both kind and extent from that usually encountered in other disciplines. Certain biocomputational methodology can be viewed as a bridge between the data forms commonly encountered in biomedicine, and the statistical and computational machinery which had previously been developed to deal with physical science information. Examples are given of three promising research subareas, all of which concern methods for dealing with highly complex forms of health and medical data.
A method is described for compacting homogeneous and separating heterogeneous data subcomponents. Allowance is made for heterogeneity due to a concomitant variable. When used with a large sample size, the major component of bivariate normal data will tend to be compacted to a single point. By applying the algorithm twice, the major component of bivariate lognormal data will tend to be compacted to a single point. These procedures have been implemented as part of an interactive graphical system for convenient use by biomedical researchers.
Editing options of an interactive graphical-biometry statistical system are described. A method for screening using a highly reduced data file is illustrated. By utilizing properties of sample Fourier coefficients, the problem of component overlap can be resolved. This, in turn, tends to free edited statistical information from the contamination or truncation inherent in the usual screening process.
Explore the source record for details and available documents.
We introduce a new form of regression that has many applications to environmental studies. For a sequence composed of key variates with prototypic value chi, this form differs from the estimation of a location parameter-based curve, mu(chi), a scale parameter-based curve, sigma(chi), or other currently used types of regression. Instead of estimating a curve location, scale, or alpha-quantile parameter, it assumes that there are two or more population subgroups; for example, consisting of unsensitized and sensitized individuals, respectively. Although within each subgroup the relationships mu(chi) or sigma(chi) may or may not be horizontal, these relationships are not deemed to be of primary importance. Instead, the mixing parameter P that indexes the proportions of the two subgroups is treated as being related to the key variate value chi. In the sense that its goal is the estimation of a proportion, the new procedure resembles logit regression. But, in terms of the continuous spectrum of values attained by the response variate, the means used to attain its goal are dissimilar from those of logit regression. Specifically, group membership is not known directly but is determined from a proxy continuous variate whose values overlap between groups. Examples are given with simulated and natural data where this new form of regression is applied. We believe that conditional switching regression is a particularly valuable research tool when chemical level chi of an induced asthma attack or birthweight chi measured in a study of the biomarker cotinine's effect on pregnancy outcomes determines whether an attack or a negative outcome occurs.(ABSTRACT TRUNCATED AT 250 WORDS)