I am trying to solve this problem:

A factory produces batteries whose duration in hours for a particular has a normal distribution with $\mu_0=53$ and ${\sigma_0}^2=25$. Now suppose that there is a change in the quality of the batteries such that of the total produced, $70$ % of them has the correct duration but the remaining $30$ % have a duration in hours with normal distribution of new parameters $\mu_1$ and $\sigma_1^2$. Let $D$ be the duration in hours of a battery chosen at random from the lot of production.

1) Calculate $\mu_1$ and ${\sigma_1}^2$ knowing that $P(D \geq 47)=0,82688$ and $P(D \geq 60)=0,05746$.

2) Calculate the density function of the duration in hours for a battery chosen at random from the lot.

I am a bit lost with the exercise, I'll write what I could do:

We have two normal distributions $X_0 \sim N(53,25)$ and $X_1 \sim N(\mu_1,\sigma_1)$, I'll call now $Z_0=\dfrac{X_0-53}{5}$ and $Z_1=\dfrac{X_1-\mu_1}{\sigma_1}$ to the standarized normal distributions, notice that $Z_0=Z_1$. I define the event $A=\{\text{a battery is not damaged}\}$. We have the following $$(*) P(D \geq 47)=P(D \geq 47 | A)P(A)+P(D \geq 47 |A^c)P(A^c)$$$$=P(X_0 \geq 47)\dfrac{7}{10}+P(X_1 \geq 47)\dfrac{3}{10}$$$$=\dfrac{7}{10}P(Z_0 \geq \dfrac{-6}{5})+\dfrac{3}{10}P(Z_1 \geq \dfrac{47-\mu_1}{\sigma_1})$$ $$=\dfrac{7}{10}F_{Z_0}(\dfrac{6}{5})+\dfrac{3}{10}P(Z_1 \geq \dfrac{47-\mu_1}{\sigma_1})$$

If I didn't mess up with the calculations, we get $$P(Z_1 \geq \dfrac{47-\mu_1}{\sigma_1})=0,6915$$

In a similar manner, we arrive to $$P(Z_1 \geq \dfrac{60-\mu_1}{\sigma_1})=0,003$$

I don't know what to do from here in order to solve for $\mu_1$ and $\sigma_1$. I would appreciate suggestions to complete the solution and to find the density function.


Actually, you are very close to the answer, what you have now are $$ \begin{align} Pr (Z_1 \ge \frac{47-\mu_1}{\sigma_1}) = 0.6915 &, Pr (Z_1 \ge \frac{60-\mu_1}{\sigma_1}) = 0.003 \end{align} $$

You just need to find $a$ and $b$ such that $Pr(Z \ge a)=0.6915$ and $Pr(Z \ge b)=0.003$ where $Z \sim N(0,1)$

Then you will have $$ \begin{align} a= \frac{47-\mu_1}{\sigma_1} & , b= \frac{60-\mu_1}{\sigma_1} \end{align} $$ which will lead you to $\mu_1$ and $\sigma_1$


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.