Scribd
Upload
20 views5 pages

AQR Quant Internship Probability Guide

Uploaded by

ronitronit685
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
20 views5 pages

AQR Quant Internship Probability Guide

Uploaded by

ronitronit685
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

JUST GIVE ME ALL PORBABLITY QUESTIONS FOR

QUANT/SDE ROLE AT AQR SUMME RINTERN WITH


ANSWERS AND CROSS QUESTIONS
Here is a comprehensive set of probability and quantitative math questions frequently asked
in AQR SDE/Quant Summer Internship interviews, each with model answers, core concepts, and
typical follow-up (cross) questions. Preparing these will position you well for any AQR or
quant/tech interview round.

1. Bayes’ Theorem
Question:
Explain Bayes’ theorem and solve a scenario where it applies.
Answer:
Bayes’ theorem updates the probability of an event based on new information:

Example:
A medical test is 99% accurate, disease rate is 1%. If a patient tests positive, what’s the
probability they have the disease?
Let D = disease, + = positive test.
, ,

So even with a positive test, there’s only a 50% chance the person has the disease, due to its
rarity.
Cross-question:
How would you update the probability if a second independent test was also positive?
Where is Bayes’ theorem used in finance?
(In updating risk models as new data comes in) [1] [2] .
2. Conditional Probability
Question:
What is conditional probability and how is it calculated?
Answer:
It’s the probability of event occurring given event has occurred:

Example:
You pick two cards, without replacement, from a deck. What’s the probability that the second is
a heart, given the first is an ace?
There are 4 aces. Probability first is ace:
If the first was a non-heart ace, 13 hearts remain; if it was ace of hearts, 12 hearts remain.
Break into cases, calculate accordingly.
Cross-questions:
What is the difference between conditional, marginal, and joint probabilities?
How does conditional probability relate to Bayes’ theorem? [3] [4] .

3. Normal Distribution and Bell Curve


Question:
What is the normal distribution? Why is it called a bell curve?
Answer:
It’s a continuous, symmetric, “bell-shaped” probability distribution with parameters mean ( )
and standard deviation ( ).
68% of values within 1 std dev, 95% within 2, 99.7% within 3 of the mean.
It models natural phenomena like heights, returns, error.
Cross-question:
Why are normal distributions important in finance?
(Because many asset returns are modeled as normal for risk estimation.)

4. Expected Value
Question:
What is expected value? Calculate the expected gain from a game.
Answer:
The expected value is the average result over many trials:
Example:
Game: Win $10 with 40% chance, lose $5 with 60%.
Expected value:

.
Cross-questions:
How is expected value used to decide if a bet or investment is good?
Can expected value ever be negative? [5] .

5. Variance and Standard Deviation


Question:
Define variance and standard deviation. Why are they important?
Answer:
Variance: Mean squared deviation from the mean.
Std Dev: Square root of variance.
They measure spread/volatility—key for quantifying risk.
Example:
Set {2, 4, 6}, mean = 4.
Variance:
Std dev: .
Cross-question:
Can you have high variance but zero mean? [5] .

6. 'The Greeks' in Options


Question:
What are “The Greeks” in options pricing? Explain delta and vega.
Answer:
Delta ( ): How much an option’s price moves for a $1 move in the underlying asset. Calls:
0–1, Puts: -1–0.
Vega: Sensitivity to volatility; how much option value changes if implied volatility moves 1%.
Gamma: Rate of delta’s change.
Theta: Rate of time decay.
Rho: Sensitivity to interest rate changes.
Cross-questions:
Why does delta for a call option approach 1?
Why might a trader care about vega exposure? [6] [7] .

7. Classic Quant Probability Puzzles


Question:
A fair die is rolled twice. What’s the probability the sum is 8?
Answer:
Ways to get sum=8: (2,6), (3,5), (4,4), (5,3), (6,2) — 5 pairs out of 36.
Probability = 5/36.
Cross-question:
If the first roll is even, what’s the chance the sum is 8?
How would you model this in code or simulation? [8] .

8. Interview Brainteasers / Scenarios


Sample Puzzles:
Biased coin:
If a coin shows heads with probability 0.7, what is the probability of 2 heads in 3 tosses?
(Use binomial: )
Monty Hall problem:
Does it benefit you to switch doors? (Yes, probability rises from 1/3 to 2/3.)
Deck problem:
From a shuffled deck, what’s the chance the cards from ace through king appear in order?
(There are 13! orderings; only 1 is strictly ace-king.)
Cross-questions:
How does independence affect probability calculation?
What is the birthday paradox and why does probability rise so quickly?
Give an example when the expected value does not match any possible outcome. [9] [10] [11]
[12] .
9. Extra Concepts and Common Interview Follows-
Ups
Probability of union/intersection of events (inclusion-exclusion principle).
Law of total probability.
Law of large numbers and central limit theorem (clt).
Difference between probability and odds.
When is conditional independence useful?
What is a martingale? Why is it important in trading theory? [13] [14] .
Mastering these core questions, example calculations, and cross-explanations—especially
around Bayes, expected value, independence, Greeks, and basic puzzles—will have you
very well prepared for any probability round or quant brainteaser at AQR or any similar
quant/tech interview.
⁂

1. [Link]
2. [Link]
3. [Link]
4. [Link]
5. [Link]
ts/expected-values-and-variance/
6. [Link]
7. [Link]
8. [Link]
411493482496-HKiB
9. [Link]
-fang-wall-street
10. [Link]
11. [Link]
12. [Link]
13. [Link]
14. [Link]

Common questions

Powered by AI

Vega represents the sensitivity of an option's price to changes in the volatility of the underlying asset. A trader would care about vega exposure because large swings in volatility can significantly impact the price of options, affecting profitability. Understanding vega exposure helps in managing volatility risk in a portfolio, allowing the trader to make informed decisions about hedging strategies or risk adjustments .

The birthday paradox states that in a group of just 23 people, there is approximately a 50% chance that at least two people share the same birthday. The probability rises quickly due to the combinatorial nature of calculating pairings within the group, highlighting the logarithmic increase in possible pair comparisons relative to the linear increase in group size. This problems demonstrates how human intuition often misjudges probability magnitudes in complex scenarios .

Variance and standard deviation measure the dispersion of returns in an investment portfolio, indicating volatility and thus risk. A high variance implies greater variability in returns, which translates to higher risk, while standard deviation provides a normalized measure of this risk. These metrics are crucial for assessing and managing the risk-return profile of a portfolio, aiding in the allocation of assets to match risk tolerance levels .

Bayes' theorem is used in finance to update the probability and thus the assessment of risks in models as new data becomes available. For example, when a financial asset's performance deviates from expected results, Bayes' theorem can adjust the likelihood of future outcomes based on the most recent performance data, thus refining risk assessments and making the model more responsive to real-world changes .

The Monty Hall problem demonstrates a counterintuitive result in probability theory through its setup, which involves choosing between remaining with an initial choice or switching after new information is revealed. It highlights the importance of conditional probability and demonstrates that switching doors increases the probability of winning from 1/3 to 2/3, contradicting initial intuition that the switch does not matter, thereby illustrating the complexity and non-intuitive results within probability and decision making .

The Law of Large Numbers states that as the number of experiments increases, the average of the results obtained should converge to the expected value, thus providing a clearer picture of the underlying probability distribution. This principle is important because it underpins the reliability of statistical estimates and models, indicating that larger samples yield more accurate reflections of true trends and probabilities .

Expected value is used to make decisions by averaging all possible outcomes of a bet or investment weighted by their probabilities. If the expected value is positive, it indicates a long-term profit potential, making it an attractive proposition. Conversely, expected value can be negative, such as a bet with a 50% chance to win $5 and a 50% chance to lose $10. The expected value is 0.5*5 + 0.5*(-10) = -2.5, suggesting a net loss over time and indicating risk .

Conditional probability is the probability of an event occurring given that another event has already occurred. For example, if you draw two cards without replacement from a deck, the probability that the second card is a heart given the first card is an ace is an instance of conditional probability. This concept is fundamental to Bayes' theorem, which uses conditional probability to update the likelihood of an event based on new information, effectively combining prior knowledge with new data .

The expected value (EV) of the game is calculated as follows: EV = (0.4 * 10) + (0.6 * -5) = 4 - 3 = 1. This means that on average, you would expect to gain $1 per game played over the long run. Although each game is independent, the expected value helps determine the overall trend of potential outcomes in multiples of such games .

The normal distribution is significant in finance because many natural phenomena, including the returns of assets, can be modeled using this distribution due to its mathematical properties of symmetry and defined probability levels within each standard deviation. It's represented as a bell curve because of its symmetrical shape, which reflects the distribution's mean-centered characteristics where most values cluster around the mean with decreasing probabilities as values move away from the mean .

You might also like