FDRestimering

FDRestimering is a statistical term referring to the process of estimating the False Discovery Rate (FDR). The False Discovery Rate is a concept used in multiple hypothesis testing, which is common in fields like genomics, neuroimaging, and econometrics where many statistical tests are performed simultaneously. When conducting many tests, there's an increased chance of obtaining false positives, meaning that a result is deemed statistically significant when it is actually due to random chance.

The FDR quantifies the expected proportion of rejected null hypotheses that are, in fact, false positives. Estimating the FDR helps researchers control the overall error rate across a set of tests. Instead of controlling the probability of making even one false positive (as with the Family-Wise Error Rate or FWER), FDR control aims to limit the proportion of false discoveries among all the discoveries made.

Several methods exist for FDR estimation, with the Benjamini-Hochberg procedure being one of the most widely used. This procedure involves ordering the p-values from the multiple tests and then determining a threshold based on the desired FDR level and the number of tests. By using FDR estimation, researchers can make more reliable interpretations of their findings when dealing with large numbers of simultaneous statistical tests, balancing the desire for discovery with the need for controlling errors.