The model assumption are satisfied in histogram, residuals vs order of collection and residuals vs fitted values.
HIstogram
We can see from the histogram that the residuals are normally distributed, so the assumption that the errors are normally distributed is met.
residuals vs. the fitted values
We find a circular pattern where the errors are concentrated based on the residuals vs. the fitted values. indicating a breach of the assumption of equal variance (homoscedasticity). There is another nonlinear pattern that should be investigated.
Residuals vs. Order of collection
: Because the residuals are randomly distributed, the assumption of equal variance of the residuals is not violated.
Residuals vs. gestational length
A The gear concentration of the residuals in the shape of an oval indicate additional signals from the data that have not been modelled. As a result, the assumption of equal variance of the residuals is violated, and it may also indicate a nonlinear relationship between the dependent and independent variables. This problem can be solved by taking the log of the independent variable and normalising the data.
Parity vs. residuals
Because this is a binary variable, the plot provides no clear indication.
Residuals vs. mother's height
There is a clear oval pattern visible, indicating that the variable is violated.
Residuals vs. mother's weight
There is a clear oval pattern visible, indicating that the variable is violated.
Smoking vs. Residuals
Because this is a binary variable, the plot provides no clear indication.
Overall, this is the next step. We must examine the variables for multicollinearity. Because the residuals histogram is normally distributed, but the residuals vs fitted values has a violation.
We must determine whether the independent variables are correlated with one another. If they are correlated, we must discard the one that is least correlated with the target variable.
To learn more about distributions refer here
https://brainly.com/question/27376566
#SPJ4
