Cheese Weight Probability via CLT Approximation

This quant interview question is about modeling the total weight of many independent cheese blocks as a sum of continuous random variables. It takes a non-normal distribution and asks you to quantify the probability that the aggregate exceeds a threshold, using a standard approximation from probability theory. In quant prep terms, it tests your ability to turn a story problem into a precise probabilistic statement involving a distributional limit.

It trains comfort with the Central Limit Theorem, continuous distributions, expectations and variances, and standardization to a familiar reference distribution. You practice moving from individual random variables to their sum, and then to a normalized quantity that can be read off from a table. It reinforces fluency with distribution parameters and scaling behavior of sums.

This matters for quant interviews because trading, risk, and derivatives pricing constantly involve approximating sums of many independent or weakly dependent risks. Interviewers use questions like this to see if you can quickly translate intuition about total exposure into a tractable probability. Strong CLT intuition is a core quant interviews skill: it powers PnL distribution approximations, VaR estimates, Monte Carlo diagnostics, and back-of-the-envelope risk calculations.

What it tests

When dealing with the sum of a large number of independent, identically distributed random variables, the Central Limit Theorem (CLT) tells us that the distribution of their sum approaches a normal distribution, regardless of the original distribution's shape, provided the variance is finite. This convergence happens because the individual irregularities of each summand average out, and the sum's distribution is governed by the aggregate mean and variance. The key is that the sum's mean is the sum of the means, and the variance is the sum of the variances, so the shape of the distribution becomes increasingly bell-shaped as the number of terms increases. The CLT is powerful because it allows us to use normal probability tools (like the standard normal CDF) to approximate probabilities for sums, even when the underlying data is not normal. This principle holds because the addition of independent random effects causes the overall distribution to smooth out, a phenomenon rooted in the law of large numbers and the properties of convolution.