Problem: You are a workers’ compensation claims analyst working for large manufacturing organization (worker’s compensation is worker injury protection; it pays medical and other expenditures for workers who were injured on job). A key part of your responsibilities is to ‘reserve’ a certain amount of funds for each workers’ compensation claim you procedure, which is essential for making sure that each grievance claim has sufficient funds to pay its costs. This needs you to forecast, or estimate the anticipated future cost of each claim, which is both time-consuming and difficult, as there are numerous factors (such as the nature of injury, worker’s age, nature of accident, and so forth.) that find out the ultimate cost of individual claims. Having a lot of historical data (past claims), you decide to use regression analysis to help you forecast the anticipated cost of the individual workers’ compensation claims.

Discuss how you would use regression to assist you with your task—specifically, describe how you would go about calculating the goodness-of-fit and the predictive efficacy of your model.