Statistical Methodology for Very Small (and Very Large) Studies скачать в хорошем качестве

Statistical Methodology for Very Small (and Very Large) Studies

Lecturer: Geert Molenberghs Using data from various clinical trials and other studies, methodology and statistical concepts are presented that remain solid also when studies are very small, such as in rare diseases, or very large. In the former case, we need to ensure that the statistical methodology leads to reliable estimates, without being riddled with computational instability. In the latter case, we need to ensure that running times remain feasible. Specific attention is devoted to the potential of surrogate markers and the omnipresent problem of incomplete data. The focus is on concepts and illustration, not on mathematical detail.