Long run event studies and the issue of overlapping and cross sectional correlation in returns
KeywordsEvent study ; Abnormal returns ; Expected returns ; Long horizon returns ; Bad - model problem ; Semi strong-form market hypothesis ; Chosen techniques ; Statistical and economics models ; Cross sectional correlation ; Stress test ; Hypothesis testing ; Level of significance ; Overlapping
This paper focuses on the description of various technical aspects related to long run event studies and the issue which leads to overlapping and cross sectional correlation in returns. The theoretical framework behind the long run event study methodology, autocorrelation and several applicable models of expected performance are further analyzed. Generally event studies serve an important purpose in capital market research as a way of testing market efficiency. Significance tests can be grouped in parametric and non parametric tests (NPTs). Parametric tests assume that individual firm's abnormal returns are normally distributed, whereas nonparametric tests do not rely on any such assumptions, the selection of the benchmark to use or the model to measure normal returns is therefore central to conduct an event study. The most common approach involves three steps: 1. Compute the parameters in the estimation period 2. Compute the forecast error (and obtain variance / covariance information for a period or over an event window, aggregate across firms and infer about the average effect 3. Regress cross – sectionally abnormal returns on relevant features of the stock supposed to influence the impact of the event. Event study is a statistical method to assess the impact of an event on the value of a firm; said event may be either typical event (earnings, investment, mergers etc) or economy wide events (inflation, interest rate, consumer confidence etc). The power of analysis which depicts unexpected return presupposes the existence of two basic rules 1. The choice of event, which is able to replicate information to the market, a type of information which has be estimated under common rules and common evaluation criteria. and 2. The choice of an estimator model capable to capture the power of unexpected odds (abnormal returns), which produced due to the event assessment.The observance of two reference rules ensure the power of the test and level of significance of results avoiding committing type I or type II errors The basic idea of event study that is being examined within the context of present paper is to find the abnormal return attributable to the event being studied by adjusting for the return that stems from the price fluctuation in sectoral level and as a whole market. As event we choice the public research of European Banking Authority regarding to capital adecancy of financial institutions, as assessment which covering more than 70% of total banking assets in the European Union. The framework of stress test provided competent authorities based on common macroeconomic scenarios, with common set of tools, including a common methodology, an internally consistent but relevant scenarios which display the quality of extination industry in adverse scenario.