Event studies – an introduction

Many students and researchers at some time deal with what is called “Event studies“. Wikipedia has the following definition: An Event study is a statistical method to assess the impact of an event on the value of a firm. Event studies can involve many types of events: Mergers & acquisitions, Initial Public Offerings, Quarterly statement announcements, etc. Starting research on specific types of events requires that you identify the information that is needed and what sources can be used to get the necessary data. Stock or Company identifiers are often very important when using several databases to get data.

Because students may not have often done an event study I recorded a 23 minute weblecture that I organised on 15 December 2011. Using a small case study some of the basic problems and solutions when doing event studies are demonstrated. For this case the databases Compustat North America, Zephyr and CRSP were used as an example.

Students at the Vrije Universiteit can find the weblecture in the Data Center course in Blackboard in the Menu-category Weblectures. The lecture is also available through iTunes U. This is the University channel available in iTunes where you can find weblectures from many universities worldwide on many subjects. The weblecture on event studies can be found and viewed in iTunes U if you search for the following string of 5 keywords:

“Event Studies introduction Vrije Universiteit

Email

Dealing with Outliers in datasets

When you are working with datasets it maye be necessary to trim or winsorize the data to remove odd or deviant numbers that are very different from the rest. Deviating numbers or values are often called Outliers. The defintion by Grubbs is: An outlying observation, or outlier, is one that appears to deviate markedly from other members of the sample in which it occurs.

One way to deal with Outliers is to Trim (= remove) data/numbers from the dataset to allow for more robust statistical analysis. Another way to deal with Outliers, is Winsorizing the data: a method of averaging that replaces the smallest and largest values with the observations closest to them. A typical Winsorizing strategy is to set all outliers to a specified percentile of the data.

The following Youtube movie explains Outliers very clearly:

If you need to deal with Outliers in a dataset you first need to find them and then you can decide to either Trim or Winsorize them. In a large dataset detecting Outliers is difficult but there are some ways this can be made easier using spreadsheet programs like Excel or SPSS. Below you can find two youtube movies for each program that shows you how to do this.

Microsoft Excel & Detecting Outliers:

Detecting Outliers using SPSS:

Email

Market research in LexisNexis

Many databases cover listed companies. When you are doing research on specific markets in countries or regions, it is also important to have some (basic) knowledge that includes unlisted companies. That way you can get a better idea of the whole market. Only a few specific databases also cover unlisted companies and offer the option to find out how many companies are active in a specific industry/market and therefore get a better idea of the whole market.

A large database that allows you to do market research is LexisNexis. This database covers very many companies worlwide. Depending on the information source or licensed module you get more or less information on each specific company. In earlier posts I told about the sources and how the interface has changed in the Academic version of LexisNexis. On the left side of your sceen it is now possible to create lists of companies:

  •  In the menu on the left side in your screen: select the option “International Companies
  •  Now click the option “Create a Company list

Using industry classification codes like SIC (Standard Industry Classification) or NAICS (North AMerican Industry Classification System) you can make your lists. If you click the “SIC Lookup” or the “NAICS Lookup” option you will go to a new screen where you can search and browse through the code systems. When you select all the codes you need you can click OK to go back to the previous screen.

After combining the search with one or more geographic locations you click the “Create” button to get a list of companies. A preview screen will appear with the top ten companies. In the top right corner a red button is available to customize the list and include some standard items. They are grouped in three categories: Contact information, Company profile, and Financial information. Example:


After selecting the information you want you can click on the button “Update Company list” to create the final list. You can export the full list to Microsoft Excel using the button “Download Selected Items“. Example:

Email

Analyst stock recommendations

Many companies, especially listed ones, are followed by business analysts. Many of these work for banks or other financial organizations that either trade on stock exchanges or they are advisors in the financial world. Based on an analysis of companies they provide recommendations on whether to buy, hold or sell stocks of these companies. Highly traded stocks, bonds, etc. are of course tracked by more analysts and more recommendations are available.

One of the sources that includes some of the analysts recommendations is Datastream. Two specific Data types are of  interest:

RECNO = Number of recommendations for an equity

RECCON = Recommendation consensus (2 decimal places).
The following values apply:

RECCON Value        Meaning
Between 1 & 1.49         Strong Buy
Between 1.5 & 2.49     Buy
Between 2.5 & 3.49     Hold
Between 3.5 & 4.49     Underperform
Between 4.5 & 5            Sell

Below you see an example search with the result. The stock of the company Aalberts Industries is followed by (on average) 10 analysts and it is part of the Euronext Amsterdam Midkap index. The stock of the company AND International Publishers is currently only followed by one analyst. The company is not part of any index on the same exchange.

Email

Using the BETA in Datastream

In an earlier post I explained how to get a historical, changing, BETA for equities from Datastream. Using the specific formula it is not that difficult. If you need to get the BETA for a larger selection of equities in one go I explained in a follow-up post how excel can be used to easily combine the formula with different equities (and even markets,) should you wish to do this.

What I may not have explained clearly is, that you can not upload the list of formulas like you would a regular upload of a list of equities. If you try to do this and then use the list the Datastream software is unable to handle this search request and will display as the end result error messages like:
$$”ER”, E200, EXPRESSION TOO LONG

or:
$$”ER”, E100, INVALID CODE OR EXPRESSION ENTERED.

The only way to use a list of equities and download a historical BETA, is using the Request Table (RT) option. Just copy the list of formulas in this and it should work. Here you see an example how the RT should look:

The download result will look as follows:

NB: A Request Table search will also allow you to use different time frames for different formulas/equities for instance if you want to do an event study using Datastream. You can download an example Request table here.

Email