It is TRUE or FALSE 1) positive autocorrelation of the residuals is better than

Question

It is TRUE or FALSE

1) positive autocorrelation of the residuals is better than negative autocorrelation of the residuals 2) If the dependent variable is negatively correlated with the independent variable, then the model is not valid and the regression should not be used 3) Point estimates contain more information than confidence interval estimates. 4) In a simple linear regression problem, the residual (error) sum of squares (SSE) will sometimes exceed the regression sum of squares (SSR) 5) In a multiple regression model, the standard error must not be greater than the level of significance for the model to be valid.

Explain please

Explanation / Answer

1. Serial correlation or autocorrelation is where error terms of a time series transfers from one period to another.A positive correlation leads to exaggerated goodness of fit and negative correlation leads to fasle statistics. But the first statement is incorrect or false. A positive correlation is not better over negative correlation.

2. Regression analysis uses the formula; Y=a+bx where Y is the dependent variable and x is the independent variable.The possibility are that the dependent variable is either positively correlated or negatively correlated with the independent variable.ex for negative correlation:Psychologists have stated that wealthy people are less satisfied with thier marriages than poor people, here wealth and marital satisfaction is negatively correlated.

Answer is false. It is possible that a dependent variable(marriage satisfaction) can be negatively correlated to independent variable( in this case wealth).

3. Point estimates are estimates that provide hypothesis test pvalues and confidence interval estimates provide much more than that including the practical importance and uncertainity of the estimate. Therefore the answer is False.

4. SSE is the vaue of the sum of squares of error and a simple linear regression model will always have the lowest total sum of SSE. SSR is the sum of the squared differences between the prediction for each observation and the population mean. Therefore the statement is True. SSE will exceed the SSR.

5. The multiple regression model allows for more than one independent variable.The error will be normally distributed, standard deviation remains constant for all values of the independent variables.

The answer is True.

Related Questions

Navigate

• Browse (All)
• Browse I
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.