Preface
‘Mathematical statistics’ never lost its attractiveness, both as a mathematical discipline and for its applications in nearly all parts of empirical research. During the last years it was found that not everything that is mathematically optimal is also practically recommendable if we are not sure whether the assumptions (for instance, normality) are valid.
As an example we consider the two‐sample t‐test that is an optimal (uniformly most powerful unbiased) test if all assumptions are fulfilled. In applications however, we are often not sure that both variances are equal. Then the approximate Welch test is preferable. Such results have been found by extensive simulation experiments that played a much greater role the last time (see the eight international conferences about this topic since 1994 under http://iws.boku.ac.at).
Therefore we wrote in 2016 a new book in German (Rasch and Schott, 2016) based on Rasch (1995) incorporating the developments of the last years.
We dropped the first part of the book from 1995 containing measure and probability theory because we have excellent books about this such as Billingsley (2012) and Kallenberg (2002).
Considering the positive resonance to this book in the community of statistics, we decided to present an English version of our book from 2016. We thank Alison Oliver for the reception into Wiley’s publishing programme.
We assume from probability theory knowledge about exponential families as well as central and non‐central ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access