150 Most Frequently Asked Questions On Quant Interviews

What is Gradient Boosting, and how does it differ fundamentally from Bagging?

What is the "kernel trick" in SVMs, and how does it map non-linear financial feature spaces into linearly separable spaces?

How do you extract hazard rates and default probabilities from observed CDS spreads?

Describe the Singleton pattern. Why is it often considered an anti-pattern in high-frequency trading systems? 150 Most Frequently Asked Questions On Quant Interviews

Master the fundamentals of probability and linear algebra before diving into complex derivative pricing models.

Even if you are applying for a pure research role, you will be asked to code. Python and C++ are the industry standards.

What does a Gamma profile look like for a long straddle option strategy? What is Gradient Boosting, and how does it

What happens to your OLS standard errors and t-statistics if your data exhibits heteroscedasticity? How do you correct for it?

: Explain margin-based loss objectives and the role of support vectors.

is positive definite, what does that imply about its eigenvalues and its determinants? Describe the Singleton pattern

Final round with the head of the desk, Priya.

The "150 Most Frequently Asked Questions" are not just a test of knowledge; they are a test of character. They measure your resilience, your ability to simplify the complex, and your speed of thought. The questions act as a filter to find those who can remain calm when the numbers are moving against them.

What is the Python GIL? How does it impact your ability to use multi-threading for CPU-bound quant research?

Describe Principal Component Analysis (PCA). How do you select the optimal number of principal components for a factor model?

blue socks. You pull out socks blindly. What is the expected number of draws to get a matching pair? How many subsets of the set do not contain any consecutive numbers?