Learning Library

← Back to Library

Monte Carlo Simulations Explained

Key Points

  • Monte Carlo simulation estimates uncertain outcomes by repeatedly sampling random variables and aggregating the results.
  • It models probabilities (e.g., dice rolls) with far fewer trials than exhaustive methods by generating many possible scenarios and averaging them.
  • The technique is popular in finance (portfolio and investment planning), risk analysis, option pricing, and extends to fields such as medicine, astrophysics, and even game‑theory puzzles like Wordle.
  • To run a Monte Carlo simulation you (1) define the predictive model’s dependent and independent variables, (2) assign probability distributions to the inputs, and (3) repeatedly sample these inputs, then assess the resulting spread using variance and standard deviation.

Full Transcript

# Monte Carlo Simulations Explained **Source:** [https://www.youtube.com/watch?v=7TqhmX92P6U](https://www.youtube.com/watch?v=7TqhmX92P6U) **Duration:** 00:04:31 ## Summary - Monte Carlo simulation estimates uncertain outcomes by repeatedly sampling random variables and aggregating the results. - It models probabilities (e.g., dice rolls) with far fewer trials than exhaustive methods by generating many possible scenarios and averaging them. - The technique is popular in finance (portfolio and investment planning), risk analysis, option pricing, and extends to fields such as medicine, astrophysics, and even game‑theory puzzles like Wordle. - To run a Monte Carlo simulation you (1) define the predictive model’s dependent and independent variables, (2) assign probability distributions to the inputs, and (3) repeatedly sample these inputs, then assess the resulting spread using variance and standard deviation. ## Sections - [00:00:00](https://www.youtube.com/watch?v=7TqhmX92P6U&t=0s) **Monte Carlo Simulations Explained** - The speaker explains how Monte Carlo simulations use random sampling to model uncertain outcomes, illustrates the method with dice rolls, and highlights their common applications in finance and investment planning. ## Full Transcript
0:00monte carlo simulation is a mathematical 0:04technique which is used to estimate the 0:06possible outcomes of an uncertain event 0:08it's a chance to see into the future and 0:12while actual time travel is still beyond 0:14us let's address three questions about 0:17monte carlo simulations to get you on 0:19your way to making better decisions 0:22come on in guys so number one 0:25how do they work 0:27multi-color simulation works by modeling 0:30the probability of different outcomes in 0:32a process or a system that cannot easily 0:35be predicted due to the intervention of 0:37random variables and it uses something 0:41called random 0:43sampling 0:46and rambling random sampling is used to 0:48generate multiple possible outcomes and 0:51calculate the average result 0:54so take for example the calculation of 0:57the probability 0:58of rolling two standard dice 1:03well if you wanted to calculate this 1:05probability the brute force way you 1:08would have to roll the dice a whole 1:11bunch say 1:1236 1:14000 times if we consider that there are 1:17six sides to a dice we have two of them 1:20and we want to run this a thousand times 1:23to get a good sample size but with a 1:27monte carlo simulation we can reduce the 1:30number of roles by randomly sampling the 1:33possible outcomes knowing there are 36 1:36combination of dice rolls and 1:38calculating the percentage of times that 1:40we get say a seven 1:42now number two 1:44who uses them there are a number of 1:47common applications for monte carlo 1:50simulations and perhaps the most well 1:52known of those is in the area 1:55of just portfolio management 1:58and also in the area of investment 2:00planning 2:01by running thousands or even millions of 2:04simulations investors can get a better 2:06idea of how their portfolio might 2:09perform under different market 2:10conditions and other common applications 2:13are things like risk analysis option 2:15pricing and planning for spare capacity 2:18but a monte carlo simulation is applied 2:20in all sorts of fields from medicine all 2:24the way through to astrophysics all the 2:27way through to figuring out 2:30what today's 2:31wordle might actually be 2:34okay number three how 2:37to run one 2:39monte carlo techniques involve three 2:41basic steps 2:43first you set up the 2:46predictive 2:48model 2:51and this is identifying both the 2:53dependent variable to be predicted and 2:56the independent variables also known as 2:58the input risk of predictive variables 3:01that will drive the predictions 3:03secondly you specify the probability 3:07distribution 3:10and that's the probability distribution 3:12of the independent variables you can use 3:14historical data or an analyst's 3:17subjective judgment to define a range of 3:20likely values and assign probability 3:22weights for each 3:23and then number three 3:26we can run 3:28simulations repeatedly generating random 3:31values of the independent variables 3:34do this until enough results are 3:37gathered to make up a representative 3:39sample of the near infinite number of 3:42possible combinations 3:44you can run as many monte carlo 3:45simulations as you wish by modifying the 3:48underlying parameters you use to 3:49simulate the data however you'll also 3:52want to compute the range of variation 3:54within a sample by calculating the 3:55variance and the standard deviation 3:57which are commonly used measures of 4:00spread 4:01the more you sample the more accurate 4:04your sampling range and then the better 4:07your estimation 4:09and while you may not be able to travel 4:10into the future with monte carlo 4:13simulation you'll have a much better 4:15idea about the possibilities 4:17that the future holds 4:20if you have any questions please drop us 4:21a line below and if you want to see more 4:24videos like this in the future please 4:26like and subscribe thanks for watching