代做STAT 464: Applied Nonparametric Statistics Final Project – Fall 2024代写留学生Matlab语言
- 首页 >> Python编程STAT 464: Applied Nonparametric Statistics
Final Project – Fall 2024
Overview
After Exam 2, I plan to cover some “additional” topics, but there will be no mandatory homework and no exams. Instead, you should devote most of your efforts during this stretch to the final project. You may choose from two types of projects:
• Application Projects: Develop a compelling research question about a dataset of interest then use nonparametric methods covered in class to answer that question. If you are a non-statistics major, then you might consider choosing a dataset from your area of study.
• Methodology Projects: There are many nonparametric methods that will not be covered before Exam 2 (see below; I hope to lecture on italicized topics after Exam 2). Study one of these topics (or another approved by me) in depth. In addition to explaining this method and its purpose, you will explore its performance through simulations. If you are interested in conducting statistics research in the future, then I encourage you to pursue this type of project.
– Multivariate Tests (see Chap. 6 of Higgins)
– Analysis of Censored Data (see Chap. 7 of Higgins)
– Multifactor Experiments (see Chap. 9 of Higgins)
– Robust model fitting (see Section 10.3 of Higgins)
– Nonparametric Bootstrap (see Chapter 8 of Higgins)
– Density Estimation (see Section 10.1 of Higgins)
– Smoothing Methods (see Section 10.2 of Higgins)
Below are some general project guidelines. Detailed criteria may be found in the rubric.
• You may work alone or with one other person.
• You must submit a paper by the last day of class (12/13). The paper should be no more than 10 pages (including figures and citations) and must be compiled from an RMarkdown or LaTeX file. To this paper, you should append cleanly-written and clearly-commented code used in your project (n.b., code does not count toward page total).
• Each individual/group will present during the last two weeks of class. I will announce the time limit once we know how many presentations there will be. I realize that those presenting in the first week will have less time to prepare than those presenting in the second week, and will consider this when assigning grades.
• You must submit a project proposal (maximum one page) by 10/25. You must submit a rough draft of your paper by 11/22.
Rubric
Written Paper (60 pts)
• Application Projects
– (10 pts) Introduction
* Presents a compelling research question and articulates the motivation for answering this question.
* Clearly describes the dataset and its relevance to the research question.
* Outlines the structure of the paper.
– (10 pts) Exploratory Analysis
* Discusses data quality (e.g., identifies missing data and outliers).
* Presents basic summary statistics and/or figures that reveal the center, spread, and/or shape of the data.
* Discusses the limitations of a parametric approach in light of these observations.
* Includes other information relevant to the dataset.
– (20 pts) Statistical Analysis
* Uses an appropriate nonparametric method to answer the research question.
* Uses an analogous parametric method to answer the research question.
* Explains any differences between the nonparametric and parametric results.
– (10 pts) Discussion
* Interprets the statistical results in context.
* Describes the impact of these results.
– (10 pts) Clarity and Organization
* The paper is free of grammatical errors and clunky wording.
* The paper is well-organized and flows with the help of transitions between paragraphs and sections.
* Figures, tables, and other non-textual content is presented professionally.
• Methodology Projects
– (10 pts) Introduction
* Describes the setting in which the chosen method should be used.
* Reviews related techniques that may be used in this setting (e.g., parametric analogues and similar nonparametric methods).
* Briefly describes the chosen method and outlines why it may be preferable to alternative techniques.
* Outlines the structure of the paper.
– (15 pts) Methodology
* Thoroughly describes the chosen method.
* Describes any relevant theoretical results.
– (15 pts) Simulation Study
* Clearly describes the various configurations of the simulation study (e.g., population distri- butions and sample sizes).
* Outlines the alternative methods included in this study (at least one must be parametric).
* Discusses the relative performance of the methods under study. Considers Type 1 error rate and power.
* Proposes explanations for the observed results.
– (10) Discussion
* Highlights the main takeaways from the paper.
* Emphasizes the advantages and disadvantages of the chosen method.
– (10) Clarity and Organization
* The paper is free of grammatical errors and clunky wording.
* The paper is well-organized and flows with the help of transitions between paragraphs and sections.
* Figures, tables, and other non-textual content is presented professionally.
Oral Presentation (30 pts)
• (15 pts) Content
– Demonstrates a deep understanding of the chosen nonparametric approach.
– Demonstrates an understanding of alternative approaches and their advantages/disadvantages.
– Application Projects: Clearly describes the research question, dataset, and analysis results.
– Methodology Projects: Clearly describes the simulation study and its findings.
• (15 pts) Communication and Organization
– Explains ideas clearly and concisely, placing an emphasis on key ideas and findings.
– Effectively uses visual aids (e.g., slides or the whiteboard). In general, it is bad to clutter visual aids with words.
– Appropriately paces the presentation, remaining within the allotted time.
– Organizes presentation logically and transitions smoothly from one section to another.
Degree of Difficulty (10 pts): You will lose points if you choose a dataset or methodology that does not challenge you.