Critical Thinking-Linear Regression Model Facilitator Critical Thinking-Linear Regression Model Linear regresssion model is a sloppy way of predicting the future because of its use of only two variables in prediction. Essentially, results about the future depend on more than two variables with disturbances and impromptu changes expected hence prediction using two variables is sloppy and risky. Obsevational studies give a range of factors that determine a situation. As much as the situation may give enough reason to believe one variable affects the situation, there are usually more forces to the changes hence the need to be extremely careful in using linear regression model in projection.
An economic example is an individual’s annual savings. Savings(Y) is a function of annual income (12X) less expenditure (Z). Y=12X-ZIn linear regression, this model is sufficient to estimate any individual’s annual savings. However, annual savings depend on many other factors. Some of the factors that may affect annual savings include emergencies, economic conditions, government regulations, interest rates in the market, and investment opportunities in the market. The regression model in the example above shows weaknesses in a number of ways.
First, it over-relies on two variables, which are likely to give erroneous information about the real economic condition. Economically, the model shows biasness in premises and analysis. The model also ignores the role of disturbances and unexpected conditions in the economic analysis. Many unexpected circumstances can affect the analysis yet their effect may be greater than the variables used for example, investment opportunities. The income of the individual may also reduce in case situations come in that requires him to remit some amount on an irregular basis.
Legal remittances like social security funds may reduce the savings amount while instantaneous earnings like bonuses and jackpots may increase the savings amount. However, the moral discipline to save plays a major role because one may earn enough to save yet their spending culture inhibits saving. All these prove that linear regression is unreliable especially in projections. The graph below demonstrates linear regression model that may be used to project the future. Relationship between expenditure and savings derived from: Rencher & Christensen, 2012ReferenceRencher, A. C., & Christensen, W. F.
(2012). Chapter 10, Multivariate regression – Section 10.1, Introduction. Methods of Multivariate Analysis Wiley Series in Probability and Statistics 709 (3rd ed. ), John Wiley & Sons, p. 19.