Hello,

Can some one let me know what the author is trying to do by comparing alpha with error -coefficient and the t-stats? **Under what condition would we reject the null?**

Thanks

Hello,

Can some one let me know what the author is trying to do by comparing alpha with error -coefficient and the t-stats? **Under what condition would we reject the null?**

Thanks

We would reject the null hypothesis if the absolute value of the *t*-statistic were greater than the *t*-critical value. Same as for any other 2-tail *t*-test.

Depending on the *α* you choose, the *t*-critical value could be around 1.6 or around 2.0 or above 2.0. In any case, 0.8 is not significant for any reasonable *α*.

Thank you for the reply.

Shouldn’t you reject the null if the T stats in higher than the critical value. As, higher t-stats means that the value is in the tail?(colored region)

What am I missing>

This is a * 2-tail* test: you can have positive excess return or negative excess return.