Question

2. At a hydrocarbon processing factory, process control involves periodic analysis of samples for a certain process quality parameter. The analytic procedure currently used is costly and time consuming. A faster and more economical alternative procedure has been proposed. However, the numbers for the quality parameter given by the alternative procedure are somewhat different from those given by the current procedure, not because of any inherent errors but because of changes in the nature of the chemical analysis. Management believes that if the numbers from the new procedure can be used to forecast reliably the corresponding numbers from the current procedure, switching to the new procedure would be reasonable and cost effective. The following data were obtained for the quality parameter by analyzing samples using both procedures:

Current (Y)	Proposed (X)	Current (Y)	Proposed (X)
3.0	3.1	3.1	3.1
3.1	3.9	2.7	2.9
3.0	3.4	3.3	3.6
3.6	4.0	3.2	4.1
3.8	3.6	2.1	2.6
2.7	3.6	3.0	3.1
2.7	3.6	2.6	2.8

a. Use linear regression to find a relation to forecast Y, which is the quality parameter from the current procedure, using the values from the proposed procedure, X.

b. Is there a strong relationship between Y and X? Explain.

Answer 1

The following steps were followed in excel.

First tabulate the data as shown below:-

The select the data and from the data tab select " Data Analysis " followed by regression as shown:-

The output of the regression is as shown below:-

The scatter plots are:-

Answer a) As per the regression the relation between the two values are Y = 0.6224 X + 0.8855

b) There is not a strong relationship between Y and X as the R^2 value is around 0.45 which shows that only 45 % of the value of Y can be explained by the value of X.