Regression in eViews

[Miro]

Hi,

I should figure out how credit ratings affect stock prices. I have 30 different stocks and credit ratings for the companies. I've been running simple linear regressions in Excel but this time the data are different. I have several columns of dependent variable values - not just one.

Can I just import file that includes 30 different rows of stock prices and another file with 30 rows of ratings in to the eViews.

If I write

ls price c rating

it should match the A column of price file with A column of rating file, B column with B column and so on.

Does it work like that? Other suggestions?

[EViews Gareth]

You would need to estimate each regression individually, or write a program to do it for you.

Or you could use the Pool object which will do all of them at the same time.

[Miro]

If I run the regressions individually then I get 30 different result tables. I could do that also in Excel. Does pool object create a single result table or can I combine 30 result tables somehow into one? If I create an index of those 30 stocks I get only one table but then the result become less significant.

[EViews Gareth]

Pool puts them into one table.

Or you could use the EqTabs add-in to collate the 30 individual resukts

[ricardo]

Hi ! A question in the test omitted variable does null hypothesis is that the coefficient of the variable evaluated is zero ?