Question

Model Assumptions:

• Assumption MLR.1 (Linear in the Parameters): The model in the population can be written as y = Bo + B1X + ... + BkXk+u where Bo, B1, ..., Bk are the unknown parameters of interest and u unobserved random error. Assumption MLR.2 (Random Sampling): We have a random samp n observations, {(Xi1, X12, ..., Xik, Yi) : 1 = 1,2,...,n}, following the population model in Assumption MLR.1. Assumption MLR.3 (No Perfect Collinearity): In the sample, none independent variables is constant, and there are no exact linear relationships among the independent variables. Show all X

Question:

Answer 1

7. the assumption MLR1 is the issue. because our model will then predict a high Apgar score from a too much weight gain and too little Apgar score in the case of too little weight gain. that's a problem. because in reality there is an optimal weight gain after which the Apgar score will start to get decreased. hence if we fit a linear model it will be a specification error because the relationship between Apgar and weight gain is not linear rather it is inverted U-shaped.

8. now assumption MLR3 will be a problem because now the dependent variable(mother's weight gain) will have the problem of multicollinearity. because we are taking the same mother for two different independent variables which may be correlated. because the mother's previous health status, her eating and exercising behaviour and hormonal profile are not likely to change in a period of two years.

Give it a thumbs up if you got some help. if you have anything to ask then feel free to ask in the comments.