JP Morgan Interview Questions By **Ranojoy Dutta**, Indian Statistical Institute

**Profile:** Model Validation Specialist

**Position **– Internship

**Process**: At first there was an online test that consisted of two papers. The first one was an MCQ assessment with 35 questions ( 5 maths, 14 probability, 11 coding MCQ, 5 calculus). The duration was 45 minutes. The second paper was a coding assessment with 2 problems. The duration was 60 minutes.

There were three panels of interviewers. I got shortlisted for two panels- Panel 2 and Panel 3. There were 2 rounds in each of these panels. Then, final selections were done based on all these rounds.

## Interview Questions

##### Round 1

**Questions asked by Panel 2**

- How would you differentiate an outlier and an anomaly?
- There are 1000 lockers numbered 1 – 1000. Suppose you open all of the lockers, then close every other locker. Then, for every third locker, you close each opened locker and open each closed locker. You follow the same pattern for every fourth locker, every fifth locker, and so on up to every thousandth locker. Which locker doors will be open when the process is complete? This problem is famously known as the 1000 locker puzzle.
- I was asked to explain my work on a particular project from my resume.
- A tire manufacturing company intends to start a business in India. I was asked to give them an estimate of the profit that they can earn in their first year. No further details were provided.

**Questions asked by Panel 3**

- Walk me through your resume.
- When do we perform Lasso?
- What are hyperparameters and why are they important?
- There are 500 white balls and 501 black balls in a jar. You can pick up as many balls as you can in one go without looking at the balls. You win if you pick up an equal number of balls, else you lose. What is the total number of balls you will pick in one go so that the chance of your winning is maximized?
- I was asked about my confidence in python.

##### Round 2

**Questions asked by Panel 2**

- Say X and Y are two standard normal random variables. Then, how do you compute P(Y<X)? Give the distributions of XY, X/Y. (Provide necessary computations).
- Give an efficient algorithm on how to find the median of a given array ( length n). Compute the time complexity.
- Given a sorted array, state an efficient algorithm to check whether the element âkâ is present in the array or not. Compute the time complexity of the algorithm.

**Questions asked by Panel 3**

- Walk me through your resume.
- What are the assumptions of OLS Regression?
- What is logistic regression? How do you measure the significance of the parameters using logistic regression?
- What is K-Means Algorithm? How do you define Gap Statistic?
- What is Autocorrelation? Explain its use.
- I was asked which language I am comfortable in to do data handling: Python or R. I answered R. Then I was given some problems on data handling which can be done using the dplyr package in R. I also had to mention alternate solutions(algorithmic approach) without using the package functions.

For any query, suggestion or feedback, please reach out to us on LinkedIn. You can find some of the resources that helped us here.

Like these JP Morgan Interview Questions, you can create an impact by talking about your interview experience. Please fill **this form** to help students get a perspective about the interview structure and questions.

You can read other articles here.

Cheers and Best!