Abstract
dc:descriptionWhich event study methods are best in non-U.S. multi-country samples? Nonparametric tests, especially the rank and generalized sign, are better specified and more powerful than common parametric tests, especially in multi-day windows. The generalized sign test is the best statistic but must be applied to buy-and-hold abnormal returns for correct specification. Market-adjusted and market-model methods with local market indexes, without conversion to a common currency, work well. The results are robust to limiting the samples to situations expected to be problematic for test specification or power. Applying the tests that perform best in simulation to merger announcements produces reasonable results.
Degree
thesis:*- Grantor dc:publisher
- Alma Mater Studiorum - Università di Bologna
- Year dc:date
- 2009
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Salotti, Valentina <1980>
- Contributors dc:contributor
-
- Cenni, Stefano
Subjects
dc:subject × 2Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- en
Identifiers
dc:identifier.*- Identifier
- urn:nbn:it:unibo-1181
- OAI identifier oai:identifier
- oai:amsdottorato.cib.unibo.it:1252