DAM Project

R Markdown

This is an R Markdown document. Markdown is a simple formatting syntax for authoring HTML, PDF, and MS Word documents. For more details on using R Markdown see http://rmarkdown.rstudio.com.

When you click the Knit button a document will be generated that includes both content as well as the output of any embedded R code chunks within the document.

0.Initial Analysis

Methodology:

• Combine data based on order ID to analyze average price, demand and revenue per customer
• Split data into COD and non-COD subsets for further analysis

1.Pricing

Analysis: Testing if statistcally significant difference exists between price paid per OrderID with/without COD payment option

Methodology:

• Compare mean prices paid with/without COD
• Test if prices are normally distributed
• Compare variances in prices with/without COD payments
• Using a suitable statistical test to compare whether there is a significant difference prices paid with/without COD payments

2. Demand

Analysis: Testing if statistcally significant difference exists between no.of products sold with/without COD payment option

Methodology:

• Compare average no.of orders paid with/without COD
• Compare variances based on number of orders with/without COD payments
• Using a non-parametric test to compare whether there is a significant difference no.of orders with/without COD payments

3.Revenue

Analysis: Testing if statistcally significant difference exists between average revenue earnt per orderID with/without COD payment option

Methodology:

• Compute revenue earnt per order ID
• Compare average revenue earnt per order ID with/without COD
• Test if revenue is normally distributed
• Compare variances in revenues earnt with/without COD payments
• Using a suitable statistical test to compare whether there is a significant difference revenue earnt with/without COD payments

4.Brands

Analysis: Analyzing effect of various brands on COD payment option and revenue earnt

Methodology:

• Create a two-way contingency table characterising the brands based on COD option and revenue earnt
• Create a correlation matrix between brands, COD and revenue earnt

5. Consumer Behavior

Analysis Areas:

• Discounts provided
• Product Type
• COD Charges
• Billing and shipping address(Tentative)
• Time of Purchase (Tentative)
• Metro v/s Non-Metro (Tentative)

Methodology:

• Study effect of vendor, website discount and their interaction on with/without COD payment (Logistic Regression)
• Study effect of product type on with/without COD payments (LR)
• Compare effect of COD charges on average revenue earnt (Appropriate Statistical Test)
• If billing and shipping address are different, is COD more/less preferred?
• Study effect order time mentioned under ‘OrderDate’ on with/without COD payment decision
• Study effect of Metro v/s Non-metro city on with/without COD payment decision by mapping

6. Regressions

Logistic Regression based on the following factors

• Total order cost
• Brand
• Vendor and Website Discount (with interaction included)
• Billing and shipping address(Tentative)
• Time of Purchase (Tentative)
• Metro v/s Non-Metro (Tentative)

========================================================