BUS 520 Trident University Regression Assumptions Discussion Create Original Post and reply to 2 class mates with reference. At least 250 words each, total

BUS 520 Trident University Regression Assumptions Discussion Create Original Post and reply to 2 class mates with reference. At least 250 words each, total 750 words.

Discussion: Regression Assumptions

Based on the text on regression assumptions and your additional research, discuss the potential impact of assumption violation on interpretation of regression results.
Is there any influence of the assumption violation on the business decision making? If so, how? If not, why?

Week 1: Provide your initial discussion post to the question. Be sure to include references to any resources you used. You should use at least one resource to help you with your initial discussion.

Week 2: Respond to at least two of your classmates’ initial posts. Your response should be substantive and further the discussion. It is OK to be critical and use critical thinking. That means you can question what your classmates say. Do they provide their own critical thinking and logic?

Classmate 1:

When mentioning the potential impact of assumption violation on interpretation of regression results a couple of statistical verbiage come into play. First, the analyst must have in mind the following items of interest: normality, equal variance and independence. Second, how to find out if these assumptions are violated and lastly, their direct effects on the analysis on hand. For instance, the x and y are correlated by a systematic linear equation and failed to do so might demonstrate biased on the statistical analysis (Myers, Well, Jr, 2013). Testing for these assumptions will result in possible answers, they are either biased and erroneous or completely satisfied. ANOVA, regression, inferential statistics all contribute and serve as diagnostic tools. Any assumptions violated on a statistical test must be repeated to enable the analyst establish correct parameters and replication. The aforesaid would guarantee statistical validity and comparison measures to the real mean.

Myers, J. L., Well, A. D., & Jr, R. F. L. (2013). Research design and statistical analysis Routledge. Retrieved from ProQuest.

Classmate 2:

Hello class,

A violation of our regression equation is caused by outliers from the date. For example, with the following data set we produce a very nice regression equation of y=1.0823x+2.049

2

5

4

7

6

9

8

11

10

13

11

15

15

17

17

18

19

20

20

22

25

29

26

35

This shows a strong correlation between a change is x effecting a change in y with an r^2 of 0.9505. Now if the sample data was changed to where the “25” was changed to a “1” it would change our formula to y=0.6835x+8.8332 and produce an r^2 of 0.3683. this one outlier completely changes our data set. There are ways to try to square, cube or square root the variable to get the data back in line and help it fit into the data set better, however the interpretation of those results can be challenging (Donovan, 2016). Another solution would be to remove the data all together, however this could be an issue if the likely hood of these events is frequent.

These events would need to be critically considered before having them influence business decisions. If a 12-year-old spends $100,000 at your store, you would need to do some investigation. Maybe it was an input error, maybe it was a kid who made it big on Twitch and just wanted to buy from your shop. In these rare cases the decision to market to all 12-year-olds, or start carrying more product, should not be taken into consideration. On the other hand, if you went from selling 3,000 unit a month, for 80 months in a row, and now you only have 2 sales because, you know… COVID-19, you would want to take that into consideration for the foreseeable future in order to reduce your overhead and stay in business. It all boils down to the underlying cause of the outlying event.

V/R

Ryan

Donovan, C. (Mar 26, 2016). Violating Regression Assumptions. Retrieved May 31, 2020 from

Module 3 – Background
Simple Linear Regression
Required Reading

The primary resource for this module is Introductory Business Statistics by Alexander, Illowsky, and Dean.

Alexander, H., Illowsky, B., & Dean, S. (2017). Introductory Business Statistics. Openstax. Retrieved from https://openstax.org/details/books/introductory-business-statistics

For Module 3, you should read through the following material in this textbook.

Chapter 13: Linear Regression and Correlation
Sections 13.1, 13.2, 13.3 and 13.7 only
This chapter introduces correlation coefficients and linear regression analysis. Section 13.7 explains how to create regression estimates in Excel. There is also tutorial link below that explains how to use that tool. We will cover the remaining sections in Module 4.

You are now familiar with several tools in the Analysis Toolpak. Regression analysis is just another one of those tools. Please review the following tutorial for help in generating regression estimates in Excel:

https://www.excel-easy.com/examples/regression.html