Monte Carlo Sim

đź“Š Investing
1

Hi, trying wrap my head round this.

Standard deviation curve has 68% in the middle with rarer chances either side. Is Monte Carlo sim the same?

300k after 20 years, is most likely 50%= 2.5m
<10% 700k and <10% 7.6m

Is that right? Thanks

2

That would be the general idea, yes.

The big caveat here is that stock returns definitely do not follow a normal distribution.

Also geometric mean is lower than arithmetic mean (they are equal when variance is 0, but for random processes the geometric mean is lower than its arithmetic counterpart by a factor proportional to the standard deviation) - expecting some 11% per year is a rather optimistic forecast.

3

Don’t forget there are 3 kind of lies: Lies, damn lies and statistics

And I’m saying this while working for THE most evil financial corporation in the world basically selling numbers and calculations.

I’m planning my retirement with a 6% average annual gain assumption (and I’d be happy with this 6% average) if it is 8, I’m better of. If it’s 11 like @Zergui says hurray! :slight_smile:

On some more serious note, an SD curve is not even used as an indicator, nor [plain]Monte Carlo sim. It’s true that there are scenarios where short of having a magic 8 ball, you can do nothing but use MCS. But you use multiple levels/scenarios/timelines per position with markov or hastings sampling strategies. Most often something like a forward looking brownian bridge is preferred to anything I wrote above.

and… even when you run these simulations for days on “cloud” muscle power and sell these numbers, I still think it’s air :slight_smile:

For most people especially if they are young, punching in numbers on a “compound interest calculator” (although we are using it for stock gains) with

  • I’m gonno save this much every month
  • I’ll increase the saving amount by this much every year
  • and I will save for 20-30 years
  • with 6% annual interest

will give a much better indication of years to come.

1 Like
4

Are you adjusting for inflation? Usually I assume 8% return and minus 3% inflation, giving 5% as growth rate in the calculator, showing value in today’s terms

5

I do make a lot of assumptions including how I’d withdraw starting from ISA and padding it with tax free withdrawals from pension + annual tax free income allowance, and how the tax-free amount will improve (surely someday? :sweat_smile:)

I find adjusting for inflation a bit difficult tho, instead I’m projecting to withdraw 2% more every year. In a way this is similar to adjusting for inflation, I just find it easier.

Subtracting a flat amount from the annual gain while I am not withdrawing is kind of difficult to get my head around (despite being somewhat fiscally literate)

6

So you’d argue the probability in real terms, is more skewed to lower quartiles.

In terms of factoring in inflation, I’ve increased living cost each year in other calcs. Just didn’t want to muddy waters here.

My plan is to have 1-3 years in cash, and withdraw to refill that pot, ideally at more optimal times than dips.

Thanks.

7

Not exactly, apologies if I gave that impression, and I don’t know how long term you were planning/projecting for.

My argument is (at least for retirement) “I cannot feed myself with statistics” so I am more inclined to plan for, lower expectations. If I project with lets say 9% and if it does not happen, when/if run I out of money I cannot say but hey the average gains were 9% I should not run out of money.

Another statistics joke here: “there goes the man who drowned in a 2-foot average depth river”