Math Problem Statement

Let X1, . . . , Xn be a random sample of size n, from a distribution with probability density

f(x) = ( 1 θ , if 2θ ≤ x ≤ 3θ, 0, elsewhere, where θ > 0. What is the maximum likelihood estimator of θ ?

Solution

To find the maximum likelihood estimator (MLE) of θ\theta, let’s go through the steps systematically.

Step 1: Write the Likelihood Function

Given that X1,X2,,XnX_1, X_2, \dots, X_n is a random sample from the distribution with the probability density function:

\frac{1}{\theta} & \text{if } 2\theta \le x \le 3\theta, \\ 0 & \text{otherwise,} \end{cases}$$ the likelihood function $$L(\theta)$$ for a sample $$X_1, X_2, \dots, X_n$$ is the product of the individual densities: $$L(\theta) = \prod_{i=1}^n f(X_i; \theta).$$ Since $$f(X_i; \theta) = \frac{1}{\theta}$$ if $$2\theta \le X_i \le 3\theta$$, we have: $$L(\theta) = \frac{1}{\theta^n} \quad \text{if } 2\theta \le X_i \le 3\theta \text{ for all } i.$$ This constraint, $$2\theta \le X_i \le 3\theta$$, for all $$i = 1, 2, \dots, n$$, implies: $$2\theta \le \min(X_i) \quad \text{and} \quad \max(X_i) \le 3\theta.$$ Rearranging these inequalities gives: $$\frac{\min(X_i)}{2} \le \theta \le \frac{\max(X_i)}{3}.$$ Therefore, the likelihood function $$L(\theta)$$ is: $$L(\theta) = \frac{1}{\theta^n} \quad \text{for } \frac{\min(X_i)}{2} \le \theta \le \frac{\max(X_i)}{3}.$$ Outside this interval, $$L(\theta) = 0$$. ### Step 2: Find the Maximum of the Likelihood Function To maximize $$L(\theta)$$, observe that $$L(\theta) = \frac{1}{\theta^n}$$ is a decreasing function of $$\theta$$. Thus, $$L(\theta)$$ is maximized when $$\theta$$ is as small as possible, within the interval: $$\frac{\min(X_i)}{2} \le \theta \le \frac{\max(X_i)}{3}.$$ The smallest possible value of $$\theta$$ in this interval is $$\theta = \frac{\max(X_i)}{3}$$. ### Conclusion The maximum likelihood estimator (MLE) of $$\theta$$ is: $$\hat{\theta} = \frac{\max(X_i)}{3}.$$ Would you like further details or have any questions about this solution? Here are five related questions you might consider: 1. How does the range of $$X_i$$ values affect the likelihood function for other distributions? 2. What are the implications of the MLE's dependency on $$\max(X_i)$$ in this case? 3. How would the MLE change if the range of $$X_i$$ were different? 4. What is the role of order statistics in finding maximum likelihood estimators? 5. How does this MLE estimator behave for small sample sizes? **Tip:** When solving MLE problems, first express the likelihood function and constraints before seeking to maximize or minimize it.

Ask a new question for Free

By Image

Drop file here or Click Here to upload

Math Problem Analysis

Mathematical Concepts

Statistics
Maximum Likelihood Estimation
Probability Density Functions
Order Statistics

Formulas

Likelihood function L(θ) = Π f(Xi; θ) for all i
f(x; θ) = 1/θ if 2θ ≤ x ≤ 3θ
MLE estimator θ̂ = max(Xi) / 3

Theorems

Maximum Likelihood Estimation
Properties of Likelihood Functions
Order Statistics

Suitable Grade Level

Undergraduate Statistics