Question

Curry Rubber manufactures rubber bands for retail companies. The accounting manager has performed a regression analysis of past data. You notice that the formula has an R-squared of 0.6, a t-value of 2.3, and a standard error of the estimate of $200,000. The estimate for next quarter costs is $2,584,072. Calculate Standard Error as a % of predicted total cost which represents a measure of precision of his regression analysis?

Answer 1

The R-squared value of .6 tells you that changes in the independent variable do not predict changes in the dependent variable very well; usually we would expect the R-squared to be above approximately 80%. The t-value of 2.3 indicates there is a reasonably reliable relationship between the independent and dependent variables. The standard error of the estimate (SE) of $200,000 on a predicted total cost of $2,584,072 indicates the range above and below this estimate within which the accountant can be relatively (67%) confident that the unknown true total cost will be. The relationship between

the SE and the predicted amount is $200,000÷$2,584,072 = 7.7%, a relatively small error percentage, which represents a favorable measure of precision.

​​​​