Linear Regression Model vs. Logistic Regression model
Both methods have their own advantages and disadvantages according to the needs. Linear regression model should be used where the data are continuously increasing or decreasing by forming on a linear basis, which forms a linear look while the logistic regression model is appropriate when data is very complicated and can’t be drawn on a linear pattern. Thus, the linear regression model is more appropriate for the less complicated data and logistic regression model is more appropriate for the much complicated data (Business Statistics, 2010).
Or, it should be expressed in the form of accuracy coefficient if there is negligible or very small difference between the accuracy coefficient of the linear regression model and logistic regression model. Then linear regression model should be used but if there is a large difference then logistic coefficient should be used.
Quantitative method for the given data set
By analysing the given data set, it can be said that the linear regression model is appropriate for the given data set. As it can say that the accuracy coefficient for linear regression model in Cost of goods is 0.9657 while of logistic regression model is 0.9939, same way, for the operating cost it is 0.8585 and 0.8839 respectively, for Interest Cost it is 0.8974 and 0.9006 respectively, and for the sales it is 0.9653 and 0.9940 respectively. As the results say the difference between the accuracy co-efficient is negligible, thus linear regression quantitative model is appropriate for the given data set.
The statistical analysis of the given data is as follows:
Dependency Problem with the data
There is a problem with the data dependency in the given data as the data is not purely dependent. The entities for which the data is provided are not fixed. There may be discrepancy in the data. The data is also very much fluctuating. Let’s look at the interest cost. The curve in the interest cost is not very fluctuating but it has no dependency.