MARKETING ANALYTICS SPRING 2024

Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit

MARKETING ANALYTICS

SPRING 2024

FINAL INDIVIDUAL ASSIGNMENT

There are 4 tasks on this assignment (20 points). The dataset is “GroceryPurchase.sav” (spss datafile).

Task 1: Data Descriptive (2 points)

Task 2: Clustering k-means (5.5 points)

Task 3: Regression Analysis (6 points)

Task 4: Logistic Regression Analysis (6.5 points)

Note: 10 points will be deducted if you do not submitspss output file. As an alternative, if you are coding in R, submit your R script.

The Problem: A grocery retailer wants to implement price promotion on products during Christmas and wants to understand purchase likelihood of shoppers. To make a solid evidence-based decision on whether or not to run a Christmas promotion campaign, the retailer entrusts the business analyst to analyzeshopper information in the retailer’s database and shopping statistics over the last year, including how shoppers responded to the Christmas promotion campaign last year. As the  business analyst entrusted with the analysis for the grocery retailer, perform the following five tasks to help understand consumer categories and responses. The dataset is named “GroceryPurchase.sav” (spss datafile) and contains demographics and past behavior of 2100 shoppers.

Data Description: Following is the description of all shopper-related variables (including continuous and categorical variables in the dataset):

Variable 1: Age: shopper’sage in years

Variable 2: Education: shopper’seducation level; 1 = "High School Graduate", 2 = "College Graduate", 3 = "Master’s Degree", 4 = "PhD"

Variable 3: Married: shopper’smarital status; 1 = “Married”, 0 = “Not married”

Variable 4: Income: shopper’s annual income in $

Variable 5: Kids: are there kids in the shopper’s household: 1 = “household has kids”, 0 = “household has no kids”

Variable 6: Membership: number of days from shopper’s membership at the grocery chain

Variable 7: Days_last: number of days from shopper’s last shopping trip

Variable 8: DealsPurchased: number of purchases made on deals by shopper in last year

Variable 9: WebPurchases: number of purchases shopper made online in last year

Variable 10: StorePurchases: number of purchases shopper made in-store in last year

Variable 11: WebVisits: number of shopper visits to grocery website in last year

Variable 12: Seafood_and_Bakery: seafood and bakery purchase tier of customer in last year; 1: “low purchase in seafood and bakery”, 2: “average purchase in seafood and bakery”, 3: “high purchase in seafood and bakery”

Variable 13: Purchase: shopper’s purchase decision in response to Christmas products promotion by grocery retailer last year; 1: purchase, 0: no purchase

Task 1: Data Descriptive (3 tasks: 2 points)

a) Discuss descriptive statistics for all continuous variables in the dataset. Also create histogram plots for all the continuous variables and discuss approximately how many observations have the highest value for each of the continuous variables. (0.5 point)

b) Check frequencies for all categorical variables in the dataset. (0.5 point)

c) Check correlations between all variables (including the continuous variables and categorical variables) in the dataset and present results in a correlation table. Discuss which variables are  significantly correlated above correlation level 0.5. (1 point)

Task 2: Clustering for Market Segmentation (4 tasks: 5.5 points)

The goal here is to segment the customers of the grocery chain based on their past behavior and demographics. Consider variables from Variable 1 to Variable 12.

a) For the pair of variables with correlation > 0.5 (as per results in Task 1:c), discuss which variable you would drop from clustering analysis and which variable would you retain for clustering the Airbnb listings. (1 point)

b) For the continuous variables, calculate the standardized variables (z-scores). In your submission, print a screenshot of the first few rows of calculated z-scores. (0.5 point)

c) Identify customer segments/clusters based on the categorical variables and the standardized continuous variables. For this purpose, run k-means clustering with 3 clusters. Do you get at least 2 good-sized clusters of customers (with sizeable number of members in at least 2 clusters)? What is  the difference in mean values of the clustering variables for the 2 sizeable clusters identified? (2 points)

d) Repeat the above task 4 with the categorical variables and unstandardized continuous variables (original values of continuous variables). Are there differences in result from the above task?

Discuss the differences, if any. (2 points)

Purchase: 1 indicates positive case (purchased in past year) and 0 indicates negative case (did not purchase in the past year).