I'm learning regressions and need to understand P>t value. Please check this attached image:

enter image description here

I know how to find the regression Coef, std error, t and confidence intervals but need to figure out how to calculate P>t values.

I've tried to look at online resources and have not been able to come up with a formula that uses the above given information. I'm not just looking for a final answer but want to understand the method/formula. The answer given for the only missing P>t value is 0.06 approx.


1 Answer 1


The p.value is the probability that the calculated statistic will be at least as "extreme" as you got (under $H_0$). Namely, if you checked: $H_0: \beta_j = 0$ with a $t$ test, that is $ t_{stat} = \frac{\hat{\beta}_j}{s.d(\hat{\beta}_j)}$, then $$ p.value = \mathbb{P}(\mathcal{T}_{(n-p)} \ge |t_{stat}|), $$
where $\mathcal{T}_{n-p}$ is student's t distribution with $n-p$ df, where $p$ is the number of coeff. (including $\beta_0$).


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .