Monte Carlo method
370
MENTIONS
65
EPISODES
45
PODCASTS
Search complete. 370 mentions across 65 episodes found for "Monte Carlo method".
Nov 5, 2026
Donna Sowa AllardHOST
Let's talk about what it's like because we love retirement so much.
Donna Sowa AllardHOST
When we do a plan, we run a Monte Carlo simulation, right? And for those of you who are not familiar with a Monte Carlo simulation, what it does is it takes your plan and runs it against about a thousand different possibilities for sequence of returns of the stock market that will average whatever you used in your plan.
Donna Sowa AllardHOST
So if you used a 5.5% rate of return, you could get that by getting 5.5% every year, and that would be my ideal way to do it.
Donna Sowa AllardHOST
But – go
Nathan BeauvaisHOST
ahead.
Nathan BeauvaisHOST
I have a good analogy that I use with folks when I'm trying to describe the Monte Carlo simulation and why sequence of return
Donna Sowa AllardHOST
matters,
Nathan BeauvaisHOST
right? And so I talk about how my sister and I were very different students growing up.
Mrdjan JankovicGUEST
Illustrate this on a vehicle experiment.
Mrdjan JankovicGUEST
I'm going to go talk about multi-agent system and holonomic robots, review some CBF-based algorithms, show Monte Carlo simulations, and then the key part of the talk is towards the end when I am explaining the results with the stability analysis.
Mrdjan JankovicGUEST
negotiations is actually this stability versus instability, and illustrate this with more complex examples than the analysis is done for.
Mrdjan JankovicGUEST
So I'm packing a lot of stuff into one slide here.
26 MINS LATER
Mrdjan JankovicGUEST
But this tells me that if I sample fast, I only need the epsilon radius margin.
Mrdjan JankovicGUEST
So here is what we compared.
Mrdjan JankovicGUEST
You're going to see Monte Carlo simulations here.
Mrdjan JankovicGUEST
I have centralized.
Scott HudsonGUEST
We'll see if it's really massive in a minute, but let's call it massive for a minute.
Scott HudsonGUEST
So what did she do? Well, what she ended up doing was to use Monte Carlo representations of the interface and a particle filtering approach to consider multiple alternatives.
Scott HudsonGUEST
Now, if you think about inputs or you think about uncertainty in interaction, you have uncertainty in the input.
Scott HudsonGUEST
You don't know what you have.
Bill KeenHOST
there's a lot of science to the financial planning world, to this modeling.
Bill KeenHOST
There's Monte Carlo simulations, statistical analysis, a lot of science to it, but there's also a lot of art to it.
Bill KeenHOST
And that's where getting to know clients and understanding their passions, where their kids lives, where their grandkids live, information about their health and their hobbies.
Bill KeenHOST
Those things are priceless when it comes to really putting a model in place, a plan in place that actually represents their lives.
12 MINS LATER
Bill KeenHOST
You're not doing both.
Bill KeenHOST
And there's a lot of science to the financial planning world, to this modeling world.
Bill KeenHOST
There's Monte Carlo simulations, statistical analysis, a lot of science to it, but there's also a lot of art to it.
Bill KeenHOST
And that's where getting to know clients and understanding their passions, where their kids lives, where their grandkids live, information about their health and their hobbies.
Frank VasquezHOST
That is why you cannot rely on what are called Gaussian distributions to really model financial markets.
Frank VasquezHOST
And those Monte Carlo simulations that are based on things like, well, here's an average return and here's a standard deviation.
Frank VasquezHOST
Yeah, that is not a good model for financial markets.
speaker_4SOUNDBITE_SPEAKER
Forget about it!
Percy LiangHOST
And also, unlike beam search, it allows us to maintain better particle diversity, especially in situations where the distribution is close to uniform.
Percy LiangHOST
So now particle filtering is also called sequential Monte Carlo.
Percy LiangHOST
And there's many, many more sophisticated extensions that I haven't covered.
Percy LiangHOST
In particular, particle filtering works for general factor graphs, not just hidden markup models.
Nolan PetersonGUEST
So each node has a price for the la-- who can deliver salt there for the cheapest, who can conceivably win it, and then that is what helps us determine where we are going to sell into the market.
Nolan PetersonGUEST
So I'll play it here 'cause it's a dynamic model, and of course it uses statistics in Monte Carlo, and you can see here this is Atlas Salt at twenty t- thirty-three, twenty thirty-four, twenty thirty-six, so on when we hit four million tons of production.
Nolan PetersonGUEST
These are the markets that we would see ourselves displacing primarily foreign imports, but there'll be some competition with other mines.
Nolan PetersonGUEST
It projects which mines may face sh- closure, closure and shutdown, whether we're there or not, uh, and certainly, of course, if we're there, there's more competition.
QuintHOST
You might be like, Quint, no, I got it, and that's fine.
QuintHOST
But if you don't, or you want a second opinion, or you want some Monte Carlo simulations, which is just a fancy word for running different scenarios and make sure that your plan doesn't break, this is why you have an advisor.
QuintHOST
Ironically, even though I was a financial advisor for 25 years...
QuintHOST
The gap I feel right now, being retired, by the way, and living off fixed income, the gap I feel is not having the administrative, quote, second pair of eyes or third pair of eyes on the overall plan.
Aswath DamodaranHOST
It might not be as high as you got when oil prices were 100, but I'm assuming that that will happen over time.
Aswath DamodaranHOST
So the question is, how do you bring the uncertainty about oil prices into your investment decision? And this is where that tool that I talked about comes into play, the Monte Carlo simulations.
Aswath DamodaranHOST
In this case, I'm very focused on one variable that I'm uncertain about, which is the oil price.
Aswath DamodaranHOST
And the nice thing about oil prices is getting a distribution of oil prices is easy to do because it's a traded commodity.
Aswath DamodaranHOST
And let me be quite honest.
Aswath DamodaranHOST
When you first look at them, the only distributions you're probably going to recognize, or maybe the only one you might recognize is a normal distribution.
Aswath DamodaranHOST
Because once you leave statistics, you forget all the rest of the stuff, right? If you're going to do Monte Carlo simulations, you might want to brush up on distributions.
Aswath DamodaranHOST
I have an eight-page write-up on my website where I take you through how to pick a distribution.
Yogesh BhattGUEST
Those are no longer relevant in the genetic world.
Yogesh BhattGUEST
And hence, you need to bring in more interesting models, like FAIR, Factor Analysis of Information Risk, or Monte Carlo Simulation, and Basel Accord Loss Distribution Approach.
Yogesh BhattGUEST
Now, traditionally, you would say these are well known in the financial space and insurance sector for risk quantification.
Yogesh BhattGUEST
And they are so much relevant at this point in time as well, because failure to deploy those kind of models is essentially going to leave a gap for CFOs to ask, hey, how many yellows before risk is red and requires immediate and expensive remediation?
Brian McKennaHOST
I mean, it's funny you mentioned Monte Carlo simulations because when I've written about them, they seem to require huge amounts of compute to do that sort of stuff.
Brian McKennaHOST
And, you know, this is the era of maybe one day when there's quantum computing, that's one of the use cases.
Brian McKennaHOST
But just moving on, moving on a little bit.
55 more episodes mention Monte Carlo method.
Create an account to see the whole feed, search across every transcript, and follow the entities you care about.