ECON 104 Lecture Notes - Lecture 9: Bayes Estimator, Loss Function

6 views2 pages
School
Department
Course
Professor

Document Summary

Quadratic loss function it looks like this, loss grows quadratically with x. In terms of money, we want to come up with how many newspapers to order, overestimate -> lose some money, underestimate -> lose consumers. Use squared error loss where a= how much we order, theta = true number of people who turns up. To determine a, they"ll come up with this value using previous observations/data. If by chance, today, more than average people turn up -> order more. Estimator comes first then look at data with estimator. We should pick what estimator to use before determining what data to use. Don"t know who much loss we will get, depends on how much to get, which depends on random sample. Risk function = expectation of loss where you take the __ over all the potential datas you might have. Estimator that minimises expected loss is a good objective, on expectation will lose least amount of money.

Get access

Grade+
$40 USD/m
Billed monthly
Grade+
Homework Help
Study Guides
Textbook Solutions
Class Notes
Textbook Notes
Booster Class
10 Verified Answers
Class+
$30 USD/m
Billed monthly
Class+
Homework Help
Study Guides
Textbook Solutions
Class Notes
Textbook Notes
Booster Class
7 Verified Answers

Related Documents

Related Questions