MAF210 Quantitative Methods for Business Trimester T1, 2025

Individual Report (Analytical)

DUE DATE AND TIME: 5 May 2025 (Monday) 8:00 PM

PERCENTAGE OF FINAL GRADE: 40% (30% (Report) +10% (Video))

HURDLE DETAILS: None

Learning Outcome Details

Instruction

This assignment is to be attempted individually. This assignment consists of FIVE parts, and each part may have multiple questions. Please answer ALL questions. The word limit for this assignment is 1300, but feel free to use more if you need it. You will not be penalised for using more words.

Written Report:

A concise, well-structured document addressing each task above. Include tables or figures where appropriate. Emphasize interpretation and clarity while you write answers to questions.

Two-Minute Video:

A clear, polished video that succinctly communicates the top insights. Format options could include a screen-recorded slideshow, a face-to-camera explanation, or a simple animated overviewwhichever best conveys the findings within the allotted time.

Submission

An electronic copy of the assignment (one copy per student including one Word document, two minute video file and one Excel file) must be uploaded to the Assignment folder under Assessment at the online unit site by 8:00 pm 5 May 2025 (Monday). The electronic copy must be named using either the student name or the student ID number. If you experience any problem in uploading the document, please contact the CloudDeakin help line. Please only upload one Word file with all questions, one video file and one Excel file with all supporting data and analysis. Please provide all explanations or answers to the questions in the Word file. Use FormulaText in the Excel file to indicate all the formulas you have used in completing this assignment.

Please note: The uploaded soft copy of your report to the unit site will be considered the official copy and it is the time of upload of this copy that will determine if the assignment was submitted on time.

Late submission

In accordance with Universitys policy, the following marking penalties will apply if you submit an assessment task after the due date without an approved extension: 5% will be deducted from available marks for each day up to five days, and work that is submitted more than five days after the due date will not be marked. You will receive 0% for the task. 'Day' means working day for paper submissions and calendar day for electronic submissions. The Unit Chair may refuse to accept a late submission where it is unreasonable or impracticable to assess the task after the due date.

Extensions can only be approved by the Unit Chair. Extensions can be granted for documented serious illness (not just on the day the assignment is due!) or for compassionate reasons under extenuating circumstances. The unit chair can ask to see how much work has been completed before granting an extension. Work or holiday reasons are NOT grounds for an extension you are expected to manage these issues as part of your studies. You are strongly encouraged to start early and to continually backup your assignment as you progress. Computer crashes or corrupted files will NOT be accepted as valid reasons for an extension of any length. For further information about Special Consideration, visit http://www.deakin.edu.au/students/assessments/special-consideration

Assessment Feedback

Students who submit their work by the due date will receive their marks and feedback within 10 business days of the due date.

Assignment: Automotive Price Analysis

Role: Junior Data Analyst at DriveData Consulting Manager: Ms. Parker (Senior Data Strategist)

Background Story

Your firm, DriveData Consulting, has just signed a contract with a client who wants a thorough analysis of what drives car prices in the market. You have been handed a dataset with 205 observations on various car models. Each observation includes the price of the vehicle and additional features. Please review the car price data.xslx and car price data dictionary.xslx to know about the features. Some of the features are in text. Please convert them to categorical variable before using. For example, the variable fueltype has two categories; gas and diesel. You can use =IF(D2="gas", 1,0). This code is appropriate if there are only two categories. Also you can use Find and replace function to convert text data to numeric data.

Ms. Parker is looking for a well-reasoned statistical report that explores (1) which features are most influential in determining price, and (2) whether certain hypotheses about differences between car groups hold true. Below is the assignment Ms. Parker has outlined.

Project Tasks

Part I: Preliminary Exploration (20 Points)

1. Forming Your Initial Conjecture

Ms. Parker wants to see your initial thoughts before you dive into the data. Write a short paragraph explaining which features you predict will have the greatest impact on Price. Provide a brief justificationwhy do you expect these features to be key drivers?

2. Data Familiarization

• Check (conceptually) for missing values or outliers. Parker wants to ensure you have a strategy in place to handle any problematic data points.

• Prepare basic summary statistics for each numerical Briefly comment on the overall picture you get from the summary statistics. You do not have to discuss the summary statistics of each variable separately.

• List any surprising or notable

Part II: Correlation Insights (20 Points)

3. Correlation Matrix

Ms. Parker is interested in a quick snapshot of how variables relate to Price.

• Create a correlation matrix between price and the

• Report which variables (top 45) show the strongest correlation with

• Provide a brief commentary on why high correlation does not automatically mean a variable causes high price. Give at least one example of a variable that might be highly correlated but not necessarily causal.

Part III : Hypothesis Testing (Variance Unknown) (20 Points)

4. Testing a Claim about a Key Feature

Ms. Parker has heard a market claim that On average, cars nowadays have at least 100 horsepower. She wants you to test whether this claim stands for the cars in your dataset, using classical hypothesis testing.

• Define the null hypothesis and the alternative hypothesis for a single population mean, recognizing that you do not know the population variance.

• Explain why the population variance is assumed

• Outline the steps you would take to compute the test statistic, degrees of freedom, and p-value.

5. Type I & Type II Errors

• Parker wants clarity on errors. Describe what a Type I error would mean in this context.

• Describe what a Type II error would mean in this

• Provide a real-world scenario showing the consequence of each type of error especially if a manufacturer is making decisions about engine performance

6. Interpretation of the Hypothesis Test

• Parker wants an example of how you would interpret a hypothetical resultfor instance, a p-value of 0.03 at ?=0.05.

• How could sample size affect your ability to detect whether the mean is truly different from 100 horsepower?

• If your final conclusion contradicts Parkers claim, how might you present this finding diplomatically?

Part IV: Multiple Linear Regression (40 Points)

7. Building the First Model

Ms. Parker requests a multiple linear regression model where price is the dependent variable and the features are potentially the explanatory variables.

• Which are the predictors do you think are relevant and why?

• Conceptually discuss the estimated coefficients, standard errors, and p-values. Which predictors appear to be statistically significant at ?=0.05?

• Do these results match the expectations you formed in Part I or the correlations in

Part II?

8. Model Diagnostics & Assumptions

Ms. Parker emphasizes model reliability. She wants you to check the classic assumptions:

• How do you verify the relationship is approximately linear ?

• Do you think the model is reliable for identifying the true relationship between price and the features you have identified from your analysis. Explain.

9. Model Evaluation

• Compare R-Square and Adjusted R-Square. Why is the Adjusted R-Square more relevant for comparing models with different numbers of predictors? Which metric do you rely on more, and why? It will be useful if you refer to the analysis you did in Q7 part (a) as you answer this question.

• Provide a one-paragraph summary answering: Is this final model good enough for the clients needs? Why or why not?

• Present two limitations that might affect results.

10. Exploring a Log-Transform

• Parker is curious if using ln(price) instead of price might improve the model. Explain how you would fit a log-linear model with the same predictors.

• Compare the interpretation of coefficients and see if the log model is

• Decide which model you recommendlinear or log-linearand why.

Part V: Final Managers Video (20 Points)

11. Two-Minute Video Presentation

Ms. Parker also wants you to create a short, 2-minute video summarizing the key takeaways of your analysis. Highlight the top 23 insights you uncovered. Think of this video as a brief briefing for upper managementkeep it concise, clear, and visually engaging if youre using slides.

Good luck with your analysis! Ms. Parker eagerly awaits your thorough yet succinct report and your two-minute video highlighting the major findings. If you need any clarifications about the tasks or guidance on classical statistical methods, reach out for an internal memo or schedule a brief meeting. Remember: Your ultimate goal is to provide actionable insights that help the client understand the key factors driving car prices.