Why is the t-statistic useful
Company: Two Sigma
Role: Data Scientist
Category: Statistics & Math
Difficulty: easy
Interview Round: Technical Screen
In classical hypothesis testing (e.g., a one-sample or two-sample mean test), explain:
## Question
1. What is the **t-statistic** and how is it computed?
2. Why is the t-statistic considered a **good measure** for testing whether an effect is real (i.e., not just noise)?
3. What assumptions are typically required for the t-test/t-statistic to be valid, and what are common failure modes?
Quick Answer: This question evaluates understanding of the t-statistic and broader statistical inference competencies, focusing on interpreting effect size relative to sampling variability in hypothesis testing.