00:01
Hello everyone, in today's question we're examining a three -variable regression model and conducting hypothesis tests to determine if each coefficient estimate is significantly different from zero.
00:12
The regression model is given with standard errors for each coefficient and an overall model standard error and r -squared value provided.
00:23
Let's test the null hypothesis hypothesis for each coefficient.
00:34
First, let's take a look at the model overview.
00:37
The model predicts y based on two independent variables, x1 and x2.
00:44
The coefficients of x1 and x2 are 0 .569 and 0 .690, respectively, with their standard errors provided as 0 .382 and 0 .288.
01:11
The intercept standard error is 1 .726.
01:22
Now the standard error for the model is 0 .991 and the r squared is 0 .684.
01:53
Hypothesis testing setup.
01:57
For each coefficient, the null hypothesis h0 states that the coefficient is equal to zero indicating no effect on that variable on y.
02:08
The t statistic for each coefficient is calculated as t equals estimate over the model's standard error of estimate.
02:31
We calculate this for the intercept x1 and x2 using the respective estimates and standard errors.
02:39
So for the intercept we have t equals negative 0 .915 divided by 1 .726...