hk1877
06-07-2008, 08:40 AM
I am testing whether it makes a difference in the valuation of companies whether they are listed on only 1 stock exchange in a country, or are also cross-listed on the NYSE (and therefore comply to a much stricter corporate governance code). In developing countries, it is often desirable to cross-list at the NYSE to signal certain standards.
I therefore have made a dataset containing all companies on a stockexchange (26 firms) of which 13 are cross-listed on NYSE. I have variables on Tobin's Q, leverage, market value and sales growth.
My regression formule looks like this:
Tobin’s Q = a + B1 [Cross-listing dummy] + B2 [log Market Value] + B3 [Leverage] + B4 [Sales growth] + nt
My first question would be how to order my data? I now have 7 columns, Firm (1-26), year (2002-2006), Tobin's Q, Cross-listing dummy, log market value and sales growth
And I have 5 rows for each company, one fore each year. Would this be the right way to sort it? My feeling is that I got confused because I need to test it for 5 years, not just 1 year.
The first thing I would like to test is whether cross-listed companies have a higher Tobin's Q rate, controlling for the other variables
The second test would be to test whether all firms performed better in 2006 (where the a certain corporate governance code was obliged) compared to 2002 (when a corporate governance code was not obligatory)
I therefore have made a dataset containing all companies on a stockexchange (26 firms) of which 13 are cross-listed on NYSE. I have variables on Tobin's Q, leverage, market value and sales growth.
My regression formule looks like this:
Tobin’s Q = a + B1 [Cross-listing dummy] + B2 [log Market Value] + B3 [Leverage] + B4 [Sales growth] + nt
My first question would be how to order my data? I now have 7 columns, Firm (1-26), year (2002-2006), Tobin's Q, Cross-listing dummy, log market value and sales growth
And I have 5 rows for each company, one fore each year. Would this be the right way to sort it? My feeling is that I got confused because I need to test it for 5 years, not just 1 year.
The first thing I would like to test is whether cross-listed companies have a higher Tobin's Q rate, controlling for the other variables
The second test would be to test whether all firms performed better in 2006 (where the a certain corporate governance code was obliged) compared to 2002 (when a corporate governance code was not obligatory)