Overview
Hi, I need help with this question with 2 parts:
This research project bridges multiple disciplines by leveraging regression analysis techniques for data-driven decision-making. Hands-on experience in applying simple, multiple, and logistic regression to solve business problems, prepares you for real-world analytical challenges in your respective fields.
Part 1: Cumulative Report (4-5 pages in APA format) and Software Files
Need the Final Report: Document findings, methodology, and recommendations in a structured format (introduction, literature review, methodology, results, discussion, and conclusions). This should be a cumulative document of Phases 1-4 of the Research Project.
Submit the Software File: The Software file with the raw data and statistical functions documented. This should be a cumulative document of the Data you collected in Phases 1-4 of the Research Project. (Please get the data from one of the references)
Part 2: Presentation with speaker notes (10 minutes)
Summarize key insights and propose actionable solutions using data visualizations and storytelling techniques. Your presentation should include audio and visual aids.
Think of your presentation like a story. Start with something interesting to grab everyone's attention in the beginning. When you talk about what others have already found, explain it like the background information of your story, showing what questions still need answers. Explain how you did your research like you're describing your plan to solve a problem. Present your findings as the exciting part where you reveal what you discovered. Finally, clearly state what your data means and what should happen next,.
-------
Descriptions of each phase are as below; Documents of each phase are attached for reference
Phase 1: Problem Definition & Data Collection
- Define a clear research problem with measurable outcomes.
- Identify relevant datasets from online repositories, company databases, or research surveys.
- Utilize a minimum of 15 scholarly resources for literature review and methodology selection.
- Deliverable: Write a one-page summary of the identified problem.
Phase 2: Data Preprocessing and Exploratory Data Analysis
The Trine Library Research Hub will assist you in finding relevant resources for your project: Trine Research Hub
- Clean and preprocess the dataset.
- Conduct EDA using statistical and visualization techniques (histograms, scatter plots, correlation analysis, etc.).
- Identify key variables and their relationships
Deliverable: Write a one-page summary of the key variables and relationship to the problem.
Phase 3:
Apply regression models (simple, multiple, logistic) to the dataset. Use software tools such as Python, R, Excel, or Google Sheets
- For each model, provide:
- The regression equation
- Interpretation of coefficients and their impact on dependent variables
- Model accuracy assessment (R-squared for simple/multiple regression, confusion matrix for logistic regression)
Deliverable: Write a one-page summary of the appropriate regression models to use in the development of the analysis.
Phase 4: Model Evaluation & Optimization
- Assess model performance using appropriate evaluation metrics (e.g., R-squared, Mean Squared Error (MSE), and confusion matrix for logistic regression).
- Optimize the model using feature selection and refinement techniques..
- Interpret results and their implications for decision-making.