Interview Prep — Question Bank

Compiled by Coding Jesus (@thecodingjesusquant and @codingjesus on YouTube)

Answers, if they exist, are abridged and may be inaccurate. Each section contains:

• Conceptual questions: Questions I’ve explicitly gotten

• Coding questions: Questions I’ve explicitly gotten

• Extras: Relevant additional questions and checks for content coverage and understanding during the preparation process

Questions may not be representative of generic interviews because they were likely tailored to my background.

Quant

Conceptual Questions

• OLS (ordinary linear regression)

  • Gauss–Markov theorem

    • What happens when its assumptions are violated? Provide examples if helpful. For example:

      • Errors have a nonzero mean

      • Errors are heteroscedastic

      • Errors exhibit correlation

  • Tower 24: We have a dataset on which we are running linear regression. What happens if we run linear regression on the following versions of the dataset?

    • Every row is duplicated. If we originally have N rows, we now have 2N.

      • Duplicating all rows significantly decreases the standard deviation of the linear-regression coefficients.

    • Every column is duplicated. If we originally have C columns, we now have 2C.

      • Duplicating a column produces an infinite number of solutions and no actual convergence.

    • What happens if only some ro