Welcome to the World of Insurance Math!

In this chapter, we are bridging the gap between pure probability and real-world insurance applications. Up until now, you’ve learned about random variables in a general sense. Now, we are going to look at how insurance companies actually use those variables to calculate how much money they expect to pay out.

Don't worry if this seems a bit abstract at first! Think of this chapter as "Probability with Rules." We take a basic loss (like a car crash) and apply rules (like deductibles and limits) to see what the final "moment" (average or variance) looks like. Let’s dive in!

1. Understanding the Two Players: Loss vs. Payment

Before we calculate moments, we have to understand what we are measuring. In Exam P, there is a big difference between the Loss and the Payment.

  • Loss (X): This is the actual amount of damage that occurred. If a tree falls on a car and causes \$5,000 in damage, \( X = 5000 \).
    • \n
  • Payment (Y): This is what the insurance company actually pays after applying policy rules. If the owner has a \$500 deductible, the company only pays \$4,500. So, \( Y = 4500 \).

Quick Tip: Always read the question carefully to see if they are asking for the moments of the Loss or the Payment. They are rarely the same!

2. The "Rules of the Game": Deductibles and Limits

In the "Univariate" (single variable) world, we usually modify our random variable \( X \) in two ways:

A. The Deductible (d)

A deductible is the amount the policyholder pays before the insurance kicks in. The payment variable \( Y \) is defined as:
\( Y = \max(0, X - d) \)

Analogy: Imagine a "height requirement" at an amusement park. If you aren't taller than the deductible \( d \), you don't get to ride (the payment is 0). If you are taller, you only get credit for the inches above the line.

B. The Policy Limit (u)

A limit is the maximum amount an insurance company will pay. The payment variable \( Y \) is defined as:
\( Y = \min(X, u) \)

Analogy: This is like an "All-You-Can-Eat" buffet with a cap. You can eat as much as you want, but once you hit the limit (say, 5 plates), the waiter stops bringing food. Even if you are still hungry (the loss is higher), the "payment" stops at \( u \).

3. Calculating the First Moment: The Expected Value (Mean)

The Expected Value \( E[Y] \) is the "long-run average" payment. Because the payment \( Y \) is a function of the loss \( X \), we use the Law of the Unconscious Statistician (LOTUS).

For a Deductible (d):

If \( X \) is a continuous random variable with a probability density function \( f(x) \):
\( E[Y] = \int_{d}^{\infty} (x - d) f(x) dx \)

Step-by-Step Logic:
1. From \( x = 0 \) to \( x = d \), the payment is 0. We ignore this part of the integral.
2. From \( x = d \) to infinity, the payment is \( x - d \).
3. Multiply the payment by the probability \( f(x) \) and integrate!

For a Policy Limit (u):

\( E[Y] = \int_{0}^{u} x f(x) dx + \int_{u}^{\infty} u f(x) dx \)

Step-by-Step Logic:
1. If the loss is less than the limit (\( 0 \) to \( u \)), the company pays the whole loss \( x \).
2. If the loss is greater than the limit (\( u \) to infinity), the company pays exactly \( u \).
3. Common Mistake: Forgetting the second part of the integral! Even when the loss is huge, the company still pays the limit \( u \). Don't leave that money on the table!

Did you know? There is a shortcut for the Mean! For any non-negative random variable:
\( E[\min(X, u)] = \int_{0}^{u} S(x) dx \)
Where \( S(x) \) is the Survival Function \( 1 - F(x) \). This is often much easier to integrate than the density function!

4. The Second Moment and Variance

Insurance companies don't just care about the average; they care about risk. Risk is measured by Variance.

The formula for Variance remains the same as you've learned previously:
\( Var(Y) = E[Y^2] - (E[Y])^2 \)

To find the second moment \( E[Y^2] \) for a deductible \( d \):
\( E[Y^2] = \int_{d}^{\infty} (x - d)^2 f(x) dx \)

Quick Review:
- First Moment = Mean \( E[Y] \)
- Second Moment = \( E[Y^2] \)
- Variance = The spread of the payments.

5. Working with Common Distributions

On Exam P, you will frequently see the Exponential Distribution in these problems because its "memoryless" property makes deductibles very easy to handle.

If \( X \sim \text{Exponential}(\theta) \):
1. The mean of the loss is \( E[X] = \theta \).
2. With a deductible \( d \), the expected payment given that a payment is made is still \( \theta \)!
3. The overall expected payment is \( E[Y] = \theta \cdot P(X > d) = \theta \cdot e^{-d/\theta} \).

Memory Aid: For the Exponential distribution, a deductible just "scales" the mean by the probability of exceeding the deductible. It's like the distribution "restarts" at the deductible point.

6. Summary of Key Takeaways

  • Loss (X) is the total damage; Payment (Y) is what the insurer pays.
  • Deductibles subtract from the loss: \( (X - d)_+ \).
  • Limits cap the loss: \( \min(X, u) \).
  • Mean Payment: Use LOTUS to integrate the payment function times the density.
  • Survival Function Shortcut: \( E[\min(X, u)] = \int_{0}^{u} S(x) dx \) is your best friend for saving time on the exam.
  • Variance: Always calculate the second moment first, then subtract the mean squared.

Final Encouragement: Integrating these piecewise functions (where the formula changes at \( d \) or \( u \)) is the hardest part of this chapter. Practice setting up the integrals for a Uniform distribution first—it's the simplest way to master the boundaries before moving on to harder calculus. You've got this!