The data was randomly generated and has the disadvantage of not being realistic natural or experimental data, but the advantage that I know precisely the correct model. The x's are in the first column and y's in the second.
(1) Find a good "linear" regression model, if possible, for predicting y from x.
(2) What is your predicted value of y, using this model, for x = 3 ?
(3) Assuming your model is correct and the errors are normally distributed, is there significant evidence that the the parameters in the model are not all zero.