Auctions. Episode 8. Baochun Li Professor Department of Electrical and Computer Engineering University of Toronto

Similar documents
Subjects: What is an auction? Auction formats. True values & known values. Relationships between auction formats

Social Network Analysis

Matching Markets and Google s Sponsored Search

Markets with Intermediaries

Markets with Intermediaries

Auctions. Agenda. Definition. Syllabus: Mansfield, chapter 15 Jehle, chapter 9

CSV 886 Social Economic and Information Networks. Lecture 4: Auctions, Matching Markets. R Ravi

Game Theory Lecture #16

Lecture 6 Applications of Static Games of Incomplete Information

ECO 426 (Market Design) - Lecture 11

Auctions. N i k o l a o s L i o n i s U n i v e r s i t y O f A t h e n s. ( R e v i s e d : J a n u a r y )

Auctions. Economics Auction Theory. Instructor: Songzi Du. Simon Fraser University. November 17, 2016

Algorithmic Game Theory

Today. Applications of NE and SPNE Auctions English Auction Second-Price Sealed-Bid Auction First-Price Sealed-Bid Auction

Preference Networks in Matching Markets

ECON Microeconomics II IRYNA DUDNYK. Auctions.

CUR 412: Game Theory and its Applications, Lecture 4

Agent and Object Technology Lab Dipartimento di Ingegneria dell Informazione Università degli Studi di Parma. Distributed and Agent Systems

Bayesian Nash Equilibrium

Auctions and Common Property

Auctions and Optimal Bidding

Chapter 17 Auctions and Bargaining. Outline. Auctions

Auctions. Microeconomics II. Auction Formats. Auction Formats. Many economic transactions are conducted through auctions treasury bills.

Notes on Auctions. Theorem 1 In a second price sealed bid auction bidding your valuation is always a weakly dominant strategy.

Auction is a commonly used way of allocating indivisible

Recap First-Price Revenue Equivalence Optimal Auctions. Auction Theory II. Lecture 19. Auction Theory II Lecture 19, Slide 1

CMSC 474, Introduction to Game Theory Introduction to Auctions

CUR 412: Game Theory and its Applications, Lecture 4

Ideal Bootstrapping and Exact Recombination: Applications to Auction Experiments

1 Theory of Auctions. 1.1 Independent Private Value Auctions

Auctions: Types and Equilibriums

CS364A: Algorithmic Game Theory Lecture #3: Myerson s Lemma

COMP/MATH 553 Algorithmic Game Theory Lecture 2: Mechanism Design Basics. Sep 8, Yang Cai

ECON 459 Game Theory. Lecture Notes Auctions. Luca Anderlini Spring 2017

March 30, Why do economists (and increasingly, engineers and computer scientists) study auctions?

Revenue Equivalence and Mechanism Design

Problem Set 3: Suggested Solutions

Game Theory and Auctions

Games of Incomplete Information ( 資訊不全賽局 ) Games of Incomplete Information

Optimal Auctions. Game Theory Course: Jackson, Leyton-Brown & Shoham

by open ascending bid ("English") auction Auctioneer raises asking price until all but one bidder drops out

Auction Theory: Some Basics

1 Intro to game theory

Game Theory. Lecture Notes By Y. Narahari. Department of Computer Science and Automation Indian Institute of Science Bangalore, India July 2012

Strategy -1- Strategic equilibrium in auctions

Bayesian games and their use in auctions. Vincent Conitzer

October An Equilibrium of the First Price Sealed Bid Auction for an Arbitrary Distribution.

CS269I: Incentives in Computer Science Lecture #14: More on Auctions

Multiunit Auctions: Package Bidding October 24, Multiunit Auctions: Package Bidding

Auctions. Book Pages Auction. Auction types. Rules to Auctions

1 Auctions. 1.1 Notation (Symmetric IPV) Independent private values setting with symmetric riskneutral buyers, no budget constraints.

Auctions Introduction

Game Theory Problem Set 4 Solutions

Auctions. Michal Jakob Agent Technology Center, Dept. of Computer Science and Engineering, FEE, Czech Technical University

Auctions. MSc Finance Theory of Finance 1: Financial Topics Autumn Arup Daripa Birkbeck College. The background

Mechanism design with correlated distributions. Michael Albert and Vincent Conitzer and

Day 3. Myerson: What s Optimal

CSE 316A: Homework 5

Topics in Contract Theory Lecture 6. Separation of Ownership and Control

Final Exam (100 Points Total)

Maximizing Winnings on Final Jeopardy!

Mechanism Design and Auctions

HW Consider the following game:

Notes for Section: Week 7

Auction types. All Pay Auction: Everyone writes down a bid in secret. The person with the highest bid wins. Everyone pays.

Strategy -1- Strategy

The Vickrey-Clarke-Groves Mechanism

Agent-Based Systems. Agent-Based Systems. Michael Rovatsos. Lecture 11 Resource Allocation 1 / 18

Maximizing Winnings on Final Jeopardy!

Probability Part #3. Expected Value

Applying Revenue Management in Auction design. A.M.A. Sman BMI-paper

Consider the following (true) preference orderings of 4 agents on 4 candidates.

EXAMPLE OF FAILURE OF EQUILIBRIUM Akerlof's market for lemons (P-R pp )

Auctions. Market Design. University of Notre Dame. Market Design (ND) Auctions 1 / 61

When we did independent private values and revenue equivalence, one of the auction types we mentioned was an all-pay auction

In the Name of God. Sharif University of Technology. Graduate School of Management and Economics

Game theory review. The English Auction How should bidders behave in the English auction?

Auctions. Michal Jakob Agent Technology Center, Dept. of Computer Science and Engineering, FEE, Czech Technical University

Auctioning one item. Tuomas Sandholm Computer Science Department Carnegie Mellon University

A simulation study of two combinatorial auctions

ECO 426 (Market Design) - Lecture 8

Parkes Auction Theory 1. Auction Theory. Jacomo Corbo. School of Engineering and Applied Science, Harvard University

ECO 426 (Market Design) - Lecture 9

Answers to Text Questions and Problems Chapter 9

Topics in Informational Economics 2 Games with Private Information and Selling Mechanisms

CSV 886 Social Economic and Information Networks. Lecture 5: Matching Markets, Sponsored Search. R Ravi

CS711 Game Theory and Mechanism Design

6.254 : Game Theory with Engineering Applications Lecture 3: Strategic Form Games - Solution Concepts

Dynamic Marginal Contribution Mechanism

The Cascade Auction A Mechanism For Deterring Collusion In Auctions

Economics 109 Practice Problems 1, Vincent Crawford, Spring 2002

University of Michigan. July 1994

Economics Honors Exam 2009 Solutions: Microeconomics, Questions 1-2

w E(Q w) w/100 E(Q w) w/

Microeconomic Theory II Preliminary Examination Solutions Exam date: June 5, 2017

Sequential and Concurrent Auction Mechanisms for Dynamic Spectrum Access

Game Theory. Jiang, Bo ( 江波 )

POSITION AUCTIONS: DESIGN, INCENTIVES AND STRATEGIES. Matr ANNO ACCADEMICO. Dipartimento di Economia e Finanza

Negotiation Master Course NEGOTIATION 9/12/09

Rules for Standard Offer Service Auction

Transcription:

Auctions Episode 8 Baochun Li Professor Department of Electrical and Computer Engineering University of Toronto

Paying Per Click 3

Paying Per Click Ads in Google s sponsored links are based on a cost-per-click model Advertisers only pay when a user actually clicks on the ad 3

Paying Per Click Ads in Google s sponsored links are based on a cost-per-click model Advertisers only pay when a user actually clicks on the ad The amount that advertisers are willing to pay per click is often surprisingly high To occupy the most prominent spot for calligraphy pens costs about $1.70 per click For some queries, the cost per click can be stratospheric $50 or more for a query on mortgage refinancing! 3

But how does a search engine set the prices per click for different queries? it would be difficult to set these prices with so many keywords!

Auctions (Chapter 9.1 9.6)

Let s first focus on a few simple types of auctions, and see how they promote different kinds of behaviour among bidders

Simple auctions Consider the case of a seller auctioning off one item to a set of buyers Assumption: a bidder has an intrinsic value for the item being auctioned She is willing to purchase the item for a price up to this value, but no higher Also called the bidder s true value for this item 7

Four main types of auctions Ascending-bid auctions (English auctions) Carried out interactively in real-time The seller gradually raise the price Bidders drop out until one bidder remains the winner at this final price 8

Four main types of auctions Descending-bid auctions (Dutch auctions) Carried out interactively in real-time Seller gradually lowers the price from some high initial value until the first moment when some bidder accepts and pays the current price 9

Four main types of auctions First-price sealed-bid auctions Bidders submit simultaneously sealed bids to the seller The highest bidder wins the object and pays the value of her bid 10

Four main types of auctions Second-price sealed-bid auctions (Vickrey auctions) Bidders submit simultaneous sealed bids to the sellers The highest bidder wins the object and pays the value of the second-highest bid William Vickrey, who proposed this type of auctions, were the first to analyze auctions with game theory (1961) 11

When are auctions appropriate? 12

When are auctions appropriate? Auctions are generally used by sellers in situations where they do not have a good estimate of the buyers true values for an item, and where buyers do not know each other s values If the intrinsic value of the buyer is known, there s no need for auctions The seller (or the buyer) simply commit to a fixed price that is just below the intrinsic value of the buyer (or just above that of the seller) 12

The goal of auctions The goal of auctions is to elicit bids from buyers that reveal these values Assuming that the buyers have independent, private, true values for the item 13

Descending-Bid and First-Price Auctions 14

Descending-Bid and First-Price Auctions In a descending-bid auction As the seller lowers the price from its high initial starting point, no bidder says anything until finally someone actually accepts the bid and pays the current price Bidders learn nothing while the auction is running, other than the fact that no one has accepted the current price yet For each bidder i, there s a first price bi at which she would be willing to break the silence and accept the item at price bi 14

Descending-Bid and First-Price Auctions In a descending-bid auction As the seller lowers the price from its high initial starting point, no bidder says anything until finally someone actually accepts the bid and pays the current price Bidders learn nothing while the auction is running, other than the fact that no one has accepted the current price yet For each bidder i, there s a first price bi at which she would be willing to break the silence and accept the item at price bi It is equivalent to a sealed-bid first-price auction: this price bi plays the role of bidder i s bid The item goes to the bidder with the highest bid value, and this bidder pays the value of her bid in exchange for the item 14

Ascending-Bid and Second-Price Auctions 15

Ascending-Bid and Second-Price Auctions In an ascending-bid auction Bidders gradually drop out as the seller steadily raises the price The winner of the auction is the last bidder remaining, and she pays the price at which the second-to-last bidder drops out For a bidder, it doesn t make sense to stay after the price exceeds her true (intrinsic and private) value Or to leave before the current price reaches her true value 15

Ascending-Bid and Second-Price Auctions In an ascending-bid auction Bidders gradually drop out as the seller steadily raises the price The winner of the auction is the last bidder remaining, and she pays the price at which the second-to-last bidder drops out For a bidder, it doesn t make sense to stay after the price exceeds her true (intrinsic and private) value Or to leave before the current price reaches her true value A bidder stays in an ascending-bid auction up to the exact moment when the current price reaches her true value The item goes to the highest bidder at a price equal to the secondhighest bid This is precisely the rule used in sealed-bid second-price auctions 15

Second-Price Auctions 16

Second-Price Auctions Main result: With independent, private values, bidding your true value is a dominant strategy in a second-price sealed-bid auction That is, the best choice of bid is exactly what the object is worth to you 16

Second-Price Auctions Main result: With independent, private values, bidding your true value is a dominant strategy in a second-price sealed-bid auction That is, the best choice of bid is exactly what the object is worth to you To show this, we need to formulate the second-price auction as a game Bidders correspond to players Let vi be bidder i s true value for the object Bidder i s strategy is an amount bi to bid as a function of her true value vi 16

Truthful bidding in second-price auctions The payoff to bidder i with value vi and bid bi is defined as follows: If bi is not the winning bid, then the payoff to i is 0. If bi is the winning bid, and some other bj is the second-place bid, then the payoff to i is vi bj. Claim: In a sealed-bid second-price auction, it is a dominant strategy for each bidder i to choose a bid bi = vi. 17

Proving the claim 18

Proving the claim We need to show that if bidder i bids bi = vi, then no deviation from this bid would improve her payoff, regardless of which strategy everyone else is using 18

Proving the claim We need to show that if bidder i bids bi = vi, then no deviation from this bid would improve her payoff, regardless of which strategy everyone else is using Two cases to consider: deviations in which i raises her bid, and deviations in which i lowers her bid 18

Proving the claim We need to show that if bidder i bids bi = vi, then no deviation from this bid would improve her payoff, regardless of which strategy everyone else is using Two cases to consider: deviations in which i raises her bid, and deviations in which i lowers her bid In both cases, the value of i s bid only affects whether i wins or loses, but it never affects how much i pays in the event that she wins which is determined entirely by the other bids 18

Deviating by raising or lowering her bid Alternate bid b i Raised bid affects outcome only if highest other bid b j is in between. If so, i wins but pays more than value. Truthful bid b i = v i Alternate bid b i Lowered bid affects outcome only if highest other bid b k is in between. If so, i loses when it was possible to win with non-negative payoff 19

First-price auctions The payoff to bidder i with value vi and bid bi is defined as follows: If bi is not the winning bid, then the payoff to i is 0. If bi is the winning bid, then the payoff to i is vi bi. Bidding your true value is no longer a dominant strategy! A payoff of 0 if you lose (as usual), and a payoff of 0 if you win, too The optimal way to bid is to shade your bid slightly downward, in order to get a positive payoff if you win If it s too close to the true value, your payoff won t be large if you win If it s too far below, you reduce the chance of winning 20