Task 1

Business roles at AgroStar require a baseline of analytical skills, and it is also critical that we are able to explain complex concepts in a simple way to a variety of audiences. This test is structured so that someone with the baseline skills needed to succeed in the role should be able to complete this in under 4 hours without assistance.

Use the data in the included sheet to address the following scenario...

Since its inception, AgroStar has been leveraging an assisted marketplace model. Given that the market potential is huge and that the target customer appreciates a physical store nearby, we have taken a call to explore the offline retail model to drive growth. The primary objective is to get a larger wallet share for AgroStar among existing customers.

Assume you are back in time, in August 2018 and you have been asked to determine the location (taluka) of the first AgroStar offline retail store.

1. What are the key factors you would use to determine the location? Why?
2. What taluka (across three states) would you look open in? Why?

Guidelines:

-- (1) Please mention any assumptions you have made and the underlying thought process -- (2) Please treat the assignment as standalone (it should be self-explanatory to someone who reads it), but we will have a follow-up discussion with you in which we will walk through your approach to this assignment. -- (3) Mention any data that may be missing that would make this study more meaningful -- (4) Kindly conduct your analysis within the spreadsheet, we would like to see the working sheet. If you face any issues due to the file size, kindly download this file and share an excel sheet with us -- (5) If you would like to append a word document/presentation to summarize, please go ahead. -- (6) In case you use any external data source/article, kindly share the source.

Task 4 Cohort

The file CDNOW_master.txt contains the entire purchase history up to the end of June 1998 of the cohort of 23,570 individuals who made their first-ever purchase at CDNOW in the first quarter of 1997. This CDNOW dataset was first used by Fader and Hardie (2001).

Each record in this file, 69,659 in total, comprises four fields: the customer's ID, the date of the transaction, the number of CDs purchased, and the dollar value of the transaction.

CustID = CDNOW_master(:,1); % customer id Date = CDNOW_master(:,2); % transaction date Quant = CDNOW_master(:,3); % number of CDs purchased Spend = CDNOW_master(:,4); % dollar value (excl. S&H)

See "Notes on the CDNOW Master Data Set" (http://brucehardie.com/notes/026/) for details of how the 1/10th systematic sample (http://brucehardie.com/datasets/CDNOW_sample.zip) used in many papers was created.

Reference:

Fader, Peter S. and Bruce G.,S. Hardie, (2001), "Forecasting Repeat Sales at CDNOW: A Case Study," Interfaces, 31 (May-June), Part 2 of 2, S94-S107.

Task 6 Zupee.csv

I have merged all three datasets into one file and also did some feature engineering. Available Data: You will be given anonymized user gameplay data in the form of 3 csv files. Fields in the data are as described below: Gameplay_Data.csv contains the following fields:

• Uid: Alphanumeric unique Id assigned to user
• Eventtime: DateTime on which user played the tournament
• Entry_Fee: Entry Fee of tournament
• Win_Loss: ‘W’ if the user won that particular tournament, ‘L’ otherwise
• Winnings: How much money the user won in the tournament (0 for ‘L’)
• Tournament_Type: Type of tournament user played (A / B / C / D)
• Num_Players: Number of players that played in this tournament

Wallet_Balance.csv contains following fields:

• Uid: Alphanumeric unique Id assigned to user
• Timestamp: DateTime at which user’s wallet balance is given
• Wallet_Balance: User’s wallet balance at given time stamp

Demographic.csv contains following fields:

• Uid: Alphanumeric unique Id assigned to user
• Installed_At: Timestamp at which user installed the app
• Connection_Type: User’s internet connection type (Ex: Cellular / Dial Up)
• Cpu_Type: Cpu type of device that the user is playing with
• Network_Type: Network type in encoded form
• Device_Manufacturer: Ex: Realme
• ISP: Internet Service Provider. Ex: Airtel
• Country
• Country_Subdivision
• City
• Postal_Code
• Language: Language that user has selected for gameplay
• Device_Name
• Device_Type

Build a basic recommendation system which is able to rank/recommend relevant tournaments and entry prices to the user. The main objectives are:

1. A user should not have to scroll too much before selecting a tournament of their preference
2. We would like the user to play as high an entry fee tournament as possible