Your paper should include an introduction in which the research questions are described; a body in which (a) methods used in the study are described, (b) the design of the simulation study is described, (c) outcomes are described, (d) results are described both in writing and with tables and/or plots, and (e) discussion is given as to the meaning and interpretation of the results; a conclusion; and an appendix wherein replicable and commented R code is presented.
Software and contributed packages used should be cited. While there is no style requirement, APA style is preferred. You may choose to use a style other than APA for your write-up so long as you are consistent throughout the paper.
This simulation paper is to be motivated by the classification simulation study described on pp. 152-154 in ISLR（see PAGE 2）. In particular, five classification methods are assessed on six data-generation scenarios based on classification accuracy (the outcome).
For your simulation, you will use only two data-generation scenarios: scenario 2 (as an example of one where the assumption of linear decision boundary is satisfied) and scenario 5 (as an example of one where the assumption of linear decision boundary is not supported).
For outcomes, you will use LR1, LDA, QDA, KNN1, and KNN10. Note that you use KNN10 instead of KNNCV because the cross-validation takes more time to run.
You will use R = 100 replications. Be sure to set a new seed for each simulation scenario. Results should be displayed with parallel boxplots, as in ISLR and the notes.
I am a seasoned writer, with many academic works credited to my name.I have read and understood the requirements of the project. I promise to do quality [login to view URL] award.