Positive predictive power

Such considerations have led to development of the concept of predictive power, which is more directly useful in evaluating the results from a screening instrument. Positive predictive power is the probability that a child identified by the instrument as having the disorder or symptom actually does have it, and negative predictive power is the probability that a child identified as not having the disorder or symptom actually does not have it. In some ways, predictive power is the converse of specificity and sensitivity, but it usually also depends on the base rate. If sensitivity is 100% then the negative predictive power would be 100%. If specificity is 100%, then positive predictive power is 100%. But when sensitivity and specificity are less than 100% (the usual situation), the base rate enters into the calculation. With 90% sensitivity, 90% specificity, and a 5% base rate, there are 4.5% true positives, 85.5%... 

These results led us to analyze the relationship between carrier-wave frequency and power density. We developed a mathematical model (6) which takes into account the changes in complex permittivity of brain tissue with frequency. This model predicted that a given electric-field intensity within a brain-tissue sample occurred at different exposure levels for 50-, 147-, and 450-MHz radiation. Using the calculated electric-field intensities in the sample as the independent variable, the model demonstrated that the RF-induced calcium-ion efflux results at one carrier frequency corresponded to those at the other frequencies for both positive and negative findings. In this paper, we present two additional experiments using 147-MHz radiation which further test both negative and positive predictions of this model. 

Equation (3), which is an application of Bayes theorem, is referred to as the Positive Predictive Value. The parameter p is unknown but believed to be very small (<0.01) for large virtual libraries. 1 - p is the power (or 1 - type II error, where ft is the false negative error rate) and a is the type I error, also called the size of a test in the hypothesis testing context, or the false positive error rate. The last equation defines the probability that a molecule is determined to be a hit in a biochemical assay given that the virtual screen predicts the molecule to be a hit. This probability is of great interest because it is valuable to have an estimate of the hit rate one can expect for a subset of molecules that are selected by a virtual screen. 

---