MHB Probability that the total annual profit is not more than 800000

  • Thread starter Thread starter mathmari
  • Start date Start date
  • Tags Tags
    Probability
mathmari
Gold Member
MHB
Messages
4,984
Reaction score
7
Hey!

An insurance office has $2500$ contracts with mean annual profit (per contract) $\mu=330$ and standard deviation $\sigma=540$. Calculate the probability that the total annual profit is not more than $800000$.

I have one the following:

The annual total profit should be not more than $800000$, that means that per contract it shoulebe not more than $\frac{800000}{2500}=320$.

So $$Z=\frac{X-\mu}{\sigma}=\frac{320-330}{540}\approx -1.31 \\ P(Z\leq -1.31)=0.0968$$ Is that the correct probability?
 
Physics news on Phys.org
Hey mathmari!

Let $n=2500$ be the number of contracts,
Let $X_i$ be the annual profit of contract $i$.
Let $X=\sum\limits_{i=1}^{n} X_i$ the annual profit.

So we want to know $P(X\not >800k)=P(X\le 800k)$.

We assume that the $X_i$ are independent and have distribution $N(\mu, \sigma^2)$.
Can we tell what the distribution of $X$ is? 🤔
 
Klaas van Aarsen said:
Let $n=2500$ be the number of contracts,
Let $X_i$ be the annual profit of contract $i$.
Let $X=\sum\limits_{i=1}^{n} X_i$ the annual profit.

So we want to know $P(X\not >800k)=P(X\le 800k)$.

We assume that the $X_i$ are independent and have distribution $N(\mu, \sigma^2)$.
Can we tell what the distribution of $X$ is? 🤔

Isn't $X$ also normal with $N(2500 \mu, 2500 \sigma^2)$ ?
 
mathmari said:
Isn't $X$ also normal with $N(2500 \mu, 2500 \sigma^2)$ ?
Yep. (Nod)

Can we calculate $P(X\le 800k)$ now?
I think it comes out slightly different than what you wrote before. 🤔
 
Klaas van Aarsen said:
Yep. (Nod)

Can we calculate $P(X\le 800k)$ now?
I think it comes out slightly different than what you wrote before. 🤔

It is $$P\left (\frac{\sum_{i=1}^{2500}X_i-2500\mu}{2500\sigma}\leq 8000000\right )$$ or not?
 
We have to apply central limit theorem, or not? :unsure:
 
mathmari said:
We have to apply central limit theorem, or not?

No need.
We already know that the sum of independent random variables with identical normal distributions is again a normal distribution.
So we don't need the $n=2500$ to be sufficiently large. 🧐

mathmari said:
It is $$P\left (\frac{\sum_{i=1}^{2500}X_i-2500\mu}{2500\sigma}\leq 8000000\right )$$ or not?

As you wrote before, the variance is $2500\sigma^2$.
So the standard deviation cannot be $2500\sigma$, can it? (Shake)

Btw, we need parentheses in the numerator.
Otherwise we have $\sum_{i=1}^{2500}X_i-2500\mu = (\sum_{i=1}^{2500}X_i)-2500\mu \ne \sum_{i=1}^{2500}(X_i-2500\mu)$. 🧐
 
Last edited:
Back
Top