Linear Regression and Scatterplot

Linear Regression and Scatterplot

Simple linear regression is the commonly used type of regression. Such is the case given its simplicity and the ease to show the relation between to variables, which is central to regression. Thus, identifying the key terms and concepts is of the essence in the understanding of how this type of regression works. Shedding light on such ideas is the central theme for this discussion.

To begin with, identifying the principle terminologies is fundamental to understanding this subject. A scatterplot is a befitting example of such terms. It refers to the area covered by the plotted data results in a graph that contains the independent variable on the x-axis and dependent variable on the y-axis (Montgomery, Peck, & Vining, 2015).

Another basic concept is the least square regression line (LSRL). It refers to the best line of fit that tries to mimic the observed values since it cannot go through all the data points when the relation is imperfect. The line minimizes the error whenever the observed values and predicted values are close enough. Thus, of significance is the fact that a uniform scatters of points above and below the regression line, implies a well-captured overall relationship (Forthofer, Lee, & Hernandez, 2014).

Lastly, the other important concept in linear progression is the correlation. It demonstrates the strength of a linear relationship. The value of the square of correlation if closer to zero indicates a weak relationship and an imperfect regression line. On the other hand, the value of correlation squared if is near to one, it indicates a close relationship and the best line of regression (Montgomery, Peck, & Vining, 2015).

In conclusion, the simple linear regression entails a lot, and an understanding of all these issues is a priority for its proper utilization. Therefore, getting conversant with these aspects is a necessity, that all must meet if interested using this method.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Reference

Forthofer, R., Lee, E., & Hernandez, M. (2014). Biostatistics: A Guide to Design, Analysis, and Discovery. Academic Press.

Montgomery, D., Peck, E., & Vining, G. (2015). Introduction to linear regression analysis (5th ed.). New York: NY: John Wiley & Sons.

(Visited 3 times, 1 visits today)