« »

## Monte Carlo Estimation

When we want to compute the average of a quantity that can take on many values, it is usually impossible to average over all possible values. Instead we can sample the possible values and determine an appropriate average. This procedure is referred to as a Monte Carlo (MC) calculation.

Program MCEstimation illustrates a simple example of a MC calculation by estimating the integral $$\int_0^1 y(x) dx = \int_0^1 \sqrt{1 - x^2} dx,\label{1}$$ where the possible values of $(x,y)$ are within a $1 \times 1$ box. The program randomly generates pairs $(x_i,y_i)$ and counts the percentage where $y_i \leq y(x_i)$. Because the area of a $1 \times 1$ box is unity, this percentage provides an estimate of the integral.

Problem: Monte Carlo integration
1. Show analytically that the integral in \eqref{1} is equal to $\pi/4$.
2. Use Program MCEstimation to estimate the integral \eqref{1} by Monte Carlo integration. Determine the error (the magnitude of the deviation from the exact answer) for trials of $n$ pairs of points equal to $n= 10^4$, $10^6$, and $10^8$. Does the error decrease with increasing $n$ on the average?
3. Estimate the integral using $n = 1000$. Repeat for a total of ten trials using different random number seeds each time. Is the magnitude of the variation of your values of the same order as the error between the average value and the exact value? For a large number of trials, the error is estimated from the standard error of the mean, which approximately equals the standard deviation divided by the square root of the number of trials.

## Resource

Problem 3.69 in Statistical and Thermal Physics: With Computer Applications, 2nd ed., Harvey Gould and Jan Tobochnik, Princeton University Press (2021).

OSP Projects: