Testing Endogeneity

[Vega Kuiper]

Dear people,

Currently I write my thesis about Risk Arbitrage in takeovers.
I need to investigate whether there is endogeneity in my model.
How can I test if their is endogeneity? I do not know which instrumental variables I need...

The independent variable of interest is the arbitrage spread, the dependent variable is takeover failure probability

In regression 2, the independent variable is dummy variable EM, indicating whether target firms are listed in emerging markets. Dependent variable is spread

Thankyou!

[Phil Bromiley]

There are many tools for this depending on what exactly you're estimating. If you search for estat endogenous, estat overid you'll see some of them. There is also user-written xtoverid for panel data. This is a pretty tricky problem - the existence of the takeover effort is also endogenous (perhaps appearing as sample selection bias) as is spread.

You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex.