# Lecture 12: Assessing and Deriving Estimators | Summary and Q&A

2.4K views
February 20, 2024
by
MIT OpenCourseWare
Lecture 12: Assessing and Deriving Estimators

## TL;DR

Estimation techniques, such as method of moments and maximum likelihood estimation, are used to derive estimators for unknown parameters in a given distribution. Various criteria, such as unbiasedness, efficiency, and consistency, can be used to evaluate the quality of an estimator.

## Questions & Answers

### Q: What is the method of moments estimation?

Method of moments involves equating population moments with sample moments. By solving the equations, we can obtain estimators for the unknown parameter(s) in the distribution.

### Q: How does maximum likelihood estimation work?

Maximum likelihood estimation involves finding the value(s) of the parameter(s) that maximize the likelihood function. This function represents the probability of observing the given data for various parameter values.

### Q: What criteria can be used to evaluate estimators?

Estimators can be evaluated based on criteria such as unbiasedness, efficiency, and consistency. Unbiasedness means that the expected value of the estimator is equal to the true parameter value. Efficiency refers to how tightly the estimator is distributed around the true parameter value. Consistency means that the estimator converges to the true parameter value as the sample size increases.

### Q: Are there other considerations besides the criteria mentioned?

Yes, other considerations can include computational simplicity, robustness to incorrect assumptions, and the specific needs of the application. Estimators may also have different properties depending on the type of data and distribution being estimated.

## Summary & Key Takeaways

• Methods of moments and maximum likelihood estimation are two commonly used techniques to derive estimators for unknown parameters in a distribution.

• Method of moments involves equating population moments with sample moments, and solving for the unknown parameter(s).

• Maximum likelihood estimation involves finding the value of the parameter(s) that maximizes the likelihood function, which represents the probability of the observed data given the parameter(s).

• Criteria such as unbiasedness, efficiency, and consistency can be used to assess the quality of an estimator.