Question

In order to estimate the gradient (slope) and vertical intercept when fitting a straight line to bivariate data, it is necessary to assume:

1		that the errors are normal and independent.
2		that the errors are dependent.
3		that the errors are autocorrelated.
4		nothing at all about the random error variable.

Answer 1

Since we just need to estimate the slope and intercept we do not need any condition on error random variables. If we have have to fit the regression model then only we need condition (1) but for estimating slope and intercept no condition or assumption are required

Correct answer is :

4) nothing at all about the random error variable.