Quant Methods (t-stat)

[travelin.man119]

I’m a little confused with when we reject Ho. I’m aware that we reject Ho if t stat > t crit or if t stat < - t crit…does - t crit just mean the critical value ended up being a negative number or they have to actually take a postiive and negative critical value. Sometimes it says reject Ho if t stat is > the absolute value of t crit. When do we take the absolute value and when does t-crit remain a negative t-crit…
please explain…thanks

 

[S2000magician]

The t-distribution is symmetric, so P(t > t-crit) = P(t < –t-crit).
If you’re doing a 2-tail test, your calculated t-statistic may be positive or it may be negative. All of the t-tables I’ve ever seen have only positive values for t-critical; if your calculated t-statistic is negative, you have to compare it to a negative t-critical value, so you find the absolute value of the t-critical value in the table, then throw a negative sign in front of it.

 

[JBrowntown]

^ Well said

 

[S2000magician]

That occurred during one of my (rare) lucid moments.