Veri Bilimi Uygulaması

Arute Solutions – Data Science Exercise

In this exercise, you are expected to forecast daily cash transactions for an ATM. The training data is provided as a .csv file. It includes 3 columns: CashIn, CashOut, and Date. CashIn column contains the total deposit amount of the ATM for the given date. CashOut column contains the withdrawal values of given date on ATM. (i.e. CashIn: Toplam yatırılan para miktarı, CashOut: Toplam çekilen para miktarı)

Using the data you have been provided, you are expected to forecast the CashIn and CashOut values between 2019-04-01 and 2019-04-30. You are free to use any model or models to conduct this prediction, as well as any preprocessing or additional features you see fit. The CashIn and CashOut values belong to the same ATM but can be treated independently, so you are basically expected to provide two separate forecasts.

At the end of this exercise, you are expected to submit the below:

  1. Your predictions for dates between 2019-04-01 – 2019-04-30 (as a predictions.csv file).
  2. Your source code for the assignment as file.
  3. The answers to the questions provided below in answers.docx file.

Challenge Details


  • You are expected to implement this exercise in Python 3. You are free to use any publicly available packages.
  • Please make sure you write your own code and provide your own answers, as we will be discussing these with you.
  • Though the expected language for your answers to the questions below is English, you can use Turkish for all or parts of the answers if you feel the need to.


1. What algorithm(s) did you use to conduct your prediction? Why did you choose it/them?
2. How did you decide for the parameters of the model you ultimately used?
3. What error metrics do you think should be used to evaluate predictions in such a problem? What error metrics should not be? Please report the error metrics that you think should be used in a subset of the data you have been provided.
4. Did you conduct any preprocessing of the data? If yes, why and how?
5. Please provide a graph of the data set you have been provided and your predictions (feel free to make your own decisions with regard to the aesthetics.)
6. What software did you use during the preparation of your answers?

Click here to download exercise data csv file.

Please submit your answers via e-mail to We’ll review your results and contact you if we find your results satisfying.

We are using cookies to give you the best experience. You can find out more about which cookies we are using or switch them off in privacy settings.
AcceptPrivacy Settings

  • Privacy Policy

Privacy Policy

This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognize you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.