Question

Help please?

You use a line of best fit for a set of data to make a prediction about an unknown value of the correlation coefficient for your data set is -0.015. how confident can you be that you're predicted value will be reasonably close to the actual value?

Answer 1

The r value is r = -0.015 which is much closer to 0 than it is to -1, so this means that the linear regression line is a bad model to estimate the points. You're better off using some other model (eg: quadratic model), or it's possible that the data isn't connected at all and it's random noise.

If r were closer to -1, then we'd be more confident that the lines all fall around the same straight line that slopes downward.

Answer 2

Explanation:

Correlation coefficient r always lying between -1 and +1 represents the strength of linear association between two variables.

When it is 0 there is no correlation, and negative sign represents negative association while positive sign represents positive association.

The nearer the value of correlation coefficient to -1 or +1 there is a strong association.

Here r = -0.015

correlation coefficient is almost equal to 0 representing the nil association between x and y.

Since there is practically no linear association the regression equation in linear form will be irrelevant. And hence we cannot be confident about the predicted value will be reasonably close to the actual value