In particular, the use of classification and regression tree (CART) methods may impose discontinuities in satellite tree-cover estimates not present in reality.
Although 3 years of data were combined, some estimates were not presented because they were unreliable, and some estimates for diagnoses, drugs, and procedures were aggregated into broader categories to attain reliability.
In the study by Wysowski et al (1987), adjusted, matched estimates were not presented, and so linear discriminant function analysis (Hand et al, 1998) was used to obtain estimates of the log odds ratio, and its standard error (ignoring the matching) from the means of E2 in cases and controls.
No study showed an overwhelming influence on the overall odds ratio estimates (data not presented).
Study site specific risk estimates were not presented in these publications.
However, adjustment for serum SHBG measured at the first visit did not change our estimates (analysis not presented).
Controlling for paternal height and weight, family history of schizophrenia, maternal and paternal age, or parity and household crowding did not change the effect estimates (results not presented).
Because replacing the HUI with each of these variables, alone and in combination, did not materially change the odds ratios, these effect estimates are not presented.
If divergence estimates were not presented, we retrieved the studied sequences and calculated divergence against the consensus sequence of the D. simulans genome.
Another study found no evidence of association with subsequent stroke risk in older women, although effect estimates were not presented to allow a full assessment of this.
