Inference on the regression parameters §

---

Notes and Ideas: §

In the Model: We learnt how to estimate the parameters and . If this
was all we wanted to do we could use least square estimates and not have to make distributional assumptions.

However, in most cases this is not enough and we want to make certain inferences on the parameters of regression:

1. We are interested in testing for the slope
2. Testing for the intercept
   1. Implication of Test : no CAUSAL relationship
      can be implied
3. Confidence interval for slope
4. Confidence interval for intercept
5. Confidence intervals and testing on the mean of Y
6. Prediction intervals ==For all this it is important to make the assumption that the error is distributed normally with mean 0 and standard deviation .==

---

Key : §

• MLR - t test and anova

---

TAGS §