Econometric Models for Industrial Organization 10033_9789813109650_tp.indd 16/11/16 10:43 AM World Scientific Lecture Notes in Economics ISSN: 2382-6118 Series Editor: Ariel Dinar (University of California, Riverside, USA) Vol 1: Financial Derivatives: Futures, Forwards, Swaps, Options, Corporate Securities, and Credit Default Swaps by George M Constantinides Vol 2: Economics of the Middle East: Development Challenges by Julia C Devlin Vol 3: Econometric Models for Industrial Organization by Matthew Shum Forthcoming: Cooperature Game Theory by Adam Brandenburger Lectures in Neuroeconomics edited by Paul Glimcher and Hilke Plassmann Herbert Moses - Econometric Models for Industrial Organization.indd 25-07-16 3:11:47 PM World Scientific Lecture Notes in Economics – Vol Econometric Models for Industrial Organization Matthew Shum Caltech World Scientific NEW JERSEY • LONDON 10033_9789813109650_tp.indd • SINGAPORE • BEIJING • SHANGHAI • HONG KONG • TAIPEI • CHENNAI • TOKYO 16/11/16 10:43 AM Published by World Scientific Publishing Co Pte Ltd Toh Tuck Link, Singapore 596224 USA office: 27 Warren Street, Suite 401-402, Hackensack, NJ 07601 UK office: 57 Shelton Street, Covent Garden, London WC2H 9HE Library of Congress Cataloging-in-Publication Data Names: Shum, Matthew, author Title: Econometric models for industrial organization / Matthew Shum (Caltech) Description: New Jersey : World Scientific, [2016] | Series: World scientific lecture notes in economics ; volume | Includes bibliographical references Identifiers: LCCN 2016030091 | ISBN 9789813109650 (hc : alk paper) Subjects: LCSH: Industrial organization (Economic theory) Econometric models Classification: LCC HD2326 S5635 2016 | DDC 338.601/5195 dc23 LC record available at https://lccn.loc.gov/2016030091 British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Copyright © 2017 by World Scientific Publishing Co Pte Ltd All rights reserved This book, or parts thereof, may not be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording or any information storage and retrieval system now known or to be invented, without written permission from the publisher For photocopying of material in this volume, please pay a copying fee through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA In this case permission to photocopy is not required from the publisher Desk Editors: Herbert Moses/Alisha Nguyen Typeset by Stallion Press Email: enquiries@stallionpress.com Printed in Singapore Herbert Moses - Econometric Models for Industrial Organization.indd 25-07-16 3:11:47 PM November 17, 2016 9:37 Econometric Models for Industrial Organization - 9in x 6in b2555-fm Preface These lecture notes were conceived and refined over a period of over 10 years, as teaching materials for a one-term course in empirical industrial organization for doctoral or masters students in economics Students should be familiar with intermediate probability and statistics, although I have attempted to make the lecture notes as self-contained as possible As lecture notes, these chapters have a breezy tone and style which I use in my classroom lectures Furthermore, I find it effective to teach otherwise technically difficult topics via close reading of representative papers Like many of the “newer” fields in economics, empirical industrial organization is better encapsulated as a canon of papers than a set of tools or models; hence commentaries as I have provided for papers in this canon may be the most useful and pedagogically efficient way to absorb the substance In any case, as lecture notes the material here is not exhaustive in any way; on the contrary, they are breezy, eclectic, and idiosyncratic — but ultimately sincere and well-intentioned Any reader who makes it through these notes should find herself upon a secure base from which she can freely pivot towards unexplored terrains As supplemental materials, I can recommend a good upper-level econometrics text, the Handbooks of Industrial Organization, and of course the research papers Good luck and have fun! v page v b2530 International Strategic Relations and China’s National Security: World at the Crossroads This page intentionally left blank b2530_FM.indd 01-Sep-16 11:03:06 AM November 17, 2016 9:37 Econometric Models for Industrial Organization - 9in x 6in b2555-fm Author’s Biography Matthew Shum received his Ph.D in Economics from Stanford University in 1998 He has taught at the University of Toronto, Johns Hopkins University, and the California Institute of Technology He currently resides in Arcadia, California with his wife and four children vii page vii b2530 International Strategic Relations and China’s National Security: World at the Crossroads This page intentionally left blank b2530_FM.indd 01-Sep-16 11:03:06 AM November 17, 2016 9:37 Econometric Models for Industrial Organization - 9in x 6in Acronyms BBL — Bajari–Benkard–Levin BLP — Berry–Levinsohn–Pakes CDF — Cumulative Distribution Function CS — Confidence Sets DO — Dynamic Optimization EDF — Empirical Distribution Function FOC — First-Order Condition FWER — Family-wise Error Rate GHK — Geweke–Hajivassiliou–Keane GMM — Generalized Method of Moments HM — Hotz–Miller LL fxn — Lorentz–Lorenz function OLS — Ordinary Least Squares ix b2555-fm page ix b2530 International Strategic Relations and China’s National Security: World at the Crossroads This page intentionally left blank b2530_FM.indd 01-Sep-16 11:03:06 AM November 17, 2016 9:38 Econometric Models for Industrial Organization - 9in x 6in b2555-ch01 Demand Estimation for Differentiated-product Markets page 19 19 ¯ Σ) Then the sample moment conditions are: θ ≡ (¯ α, β, se only mm,J (θ) ≡ J J ¯ pj Zmj , δj − Xj β¯ + α j=1 and we estimate θ by minimizing a quadratic norm in these sample moment functions: QJ (θ) ≡ [mm,J (θ)]m WJ [mm,J (θ)]m , θ WJ is a (M × M )-dimensional weighting matrix But problem is that we cannot perform inversion step as before, so that we cannot derive δ1 , , δJ So BLP propose a “nested” estimation algorithm, with an “inner loop” nested within an “outer loop.” • In the outer loop, we iterate over different values of the parameters Let θˆ be the current values of the parameters being considered ˆ we wish to • In the inner loop, for the given parameter values θ, ˆ In order to this, we must: evaluate the objective function Q(θ) ˆ , δJ (θ) ˆ to ˆ we solve for the mean utilities δ1 (θ), At current θ, solve the system of equations δ1 , , δJ ; θˆ sˆ1 = s˜RC δ1 , , δJ ; θˆ sˆJ = s˜RC J Note that, since we take the parameters θˆ as given, this system ˆ , δJ (θ) ˆ is J equations in the J unknowns δ1 (θ), ˆ ˆ For the resulting δ1 (θ), , δJ (θ), calculate ˆ WJ [mm,J (θ)] ˆ m ˆ = [mm,J (θ)] Q(θ) m (1.5) • Then we return to the outer loop, which searches until it finds parameter values θˆ which minimize Eq (1.5) November 17, 2016 20 9:38 Econometric Models for Industrial Organization - 9in x 6in b2555-ch01 Econometric Models for Industrial Organization • Essentially, the original inversion step is now nested inside of the estimation routine se only Note that, typically, for identification, a necessary condition is that: M = dim(Z) ≥ dim(θ) > dim(α, β) = dim([X, p]) This is because there are coefficients Σ associated with the distribution of random coefficients This implies that, even if there were no price endogeneity problem, so that (X, p) are valid instruments, we still need additional instruments in order to identify the additional parameters.1 Within this nested estimation procedure, we can also add a supply side to the RC model With both demand and supply-side moment conditions, the objective function becomes: Q (θ, γ) = GJ (θ, γ) WJ GJ (θ, γ), where GJ is the (M + N )-dimensional vector of stacked sample moment conditions: J ¯ (θ) − X β + α ¯ p δ z j j j 1j J j=1 1 J ¯ ¯ pj zM j J j=1 δj (θ) − Xj β + α , GJ (θ, γ) ≡ J (c (θ) − w γ) u j j 1j j=1 J J j=1 (cj (θ) − wj γ) uN j J where M is the number of demand side IV’s, and N the number of supply-side IV’s (Assuming M + N ≥ dim(θ) + dim(γ).) The only change in the estimation routine described in the previous section is that the inner loop is more complicated: In the inner loop, for the given parameter values θˆ and γˆ , we ˆ γˆ ) In order to this wish to evaluate the objective function Q(θ, See Moon, Shum, and Weidner (2012) page 20 November 17, 2016 9:38 Econometric Models for Industrial Organization - 9in x 6in b2555-ch01 Demand Estimation for Differentiated-product Markets page 21 21 we must: se only ˆ , δJ (θ) ˆ as ˆ solve for the mean utilities δ1 (θ), At current θ, previously ˆ , δJ (θ), ˆ calculate For the resulting δ1 (θ), ˆ ˆ ˆ ˜RC ˜RC s˜RC j (θ) ≡ s (δ(θ)), , s J (δ(θ)) and also the partial derivative matrix RC ˆ ˆ ∂˜ sRC ∂˜ s1 (δ(θ)) (δ(θ)) ∂p1 ∂p2 ∂˜ RC ˆ ˆ ∂˜ sRC s2 (δ(θ)) (δ(θ)) ∂p1 ∂p2 ˆ = D(θ) ∂˜ ˆ ˆ ∂˜ sRC sRC J (δ(θ)) J (δ(θ)) ∂p1 ∂p2 ··· ··· ··· , ˆ ∂˜ sRC (δ(θ)) ∂pJ RC ˆ ∂˜ s2 (δ(θ)) ∂pJ RC ˆ ∂˜ sJ (δ(θ)) ∂pJ For MNL case, these derivatives are: ∂sj = ∂pk −αsj (1 − sj ) −αsj sk for j = k for j = k Use the supply-side best response equations to solve for ˆ , cJ (θ): ˆ c1 (θ), p1 − c1 ˆ ˆ s˜RC = j (θ) + D(θ) ∗ pJ − cJ ˆ γˆ ) So now, you can compute G(θ, 1.5.1 Simulating the integral in Eq (1.4) The principle of simulation: approximate an expectation as a sample average Validity is ensured by law of large numbers November 17, 2016 22 9:38 Econometric Models for Industrial Organization - 9in x 6in b2555-ch01 Econometric Models for Industrial Organization In the case of Eq (1.4), note that in the integral there is an expectation: se only ¯ Σ E α ¯ , β, ≡ EG ¯ j − (αi − α ¯ )pj exp δj + (βi − β)X ¯ Σ , |¯ α, β, J ¯ j − (αi − α exp δj + (βi − β)X ¯ )pj 1+ j =1 where the random variables are αi and βi , which we assume to be ¯ ,Σ drawn from the multivariate normal distribution N (¯ α, β) For s = 1, , S simulation draws: Draw us1 , us2 independently from N(0,1) ˆ ¯ Σ, ˆ transform (us , us ) ˆ For the current parameter estimates α ¯ , β, ˆ ¯ , Σ) ˆ using the transformation ˆ into a draw from N ((α ¯ , β) αs βs = ˆ α ¯ ¯ βˆ ˆ 1/2 +Σ us1 , us2 ˆ 1/2 is shorthand for the “Cholesky factorization” of the where Σ ˆ The Cholesky factorization of a square symmetric matrix Σ matrix Γ is the triangular matrix G such that G G = Γ, so roughly it can be thought of as a matrix-analogue of “square root.” ˆ 1/2 We use the lower triangular version of Σ Then approximate the integral by the sample average (over all the simulation draws) ˆ ¯ Σ ˆ ≈ ˆ E α ¯ , β, S S s=1 1+ ¯ Xj − αs − α ˆ exp δj + β s − βˆ ¯ pj J ˆ s−β s−α ¯ ˆ exp δ + β X − α ¯ p j j j j =1 ˆ ¯ Σ, ˆ the law of large numbers ensure that this approxiˆ For given α ¯ , β, mation is accurate as S → ∞ (Results: Marginal costs and markups from BLP paper.) 1.6 Applications Applications of this methodology have been voluminous Here we discuss just a few page 22 November 17, 2016 9:38 Econometric Models for Industrial Organization - 9in x 6in b2555-ch01 se only Demand Estimation for Differentiated-product Markets page 23 23 Evaluation of voluntary export restraints (VERs): In Berry, Levinsohn, and Pakes (1999), this methodology is applied to evaluate the effects of VERs These were voluntary quotas that the Japanese auto manufacturers abided by and which restricted their exports to the United States during the 1980’s The VERs not affect the demand-side, but only the supplyside Namely, firm profits are given by: πk = (pj − cj − λVERk )Dj j∈K In the above, VERk are dummy variables for whether firm k is subject to VER (so whether firm k is Japanese firm) VER is modeled as an “implicit tax,” with λ ≥ functioning as a per-unit tax: if λ = 0, then the VER has no effect on behavior, while λ > implies that VER is having an effect similar to increase in marginal cost cj The coefficient λ is an additional parameter to be estimated, on the supply-side (Results: Effects of VER on firm profits and consumer welfare.) Welfare from new goods, and merger evaluation: After cost function parameters γ are estimated, you can simulate equilibrium prices under alternative market structures, such as mergers, or entry (or exit) of firms or goods These counterfactual prices are valid assuming that consumer preferences and firms’ cost functions don’t change as market structures change Petrin (2002) presents consumer welfare benefits from introduction of the minivan, and Nevo (2001) presents merger simulation results for the ready-to-eat cereal industry Geographic differentiation: In our description of BLP model, we assume that all consumer heterogeneity is unobserved Some models have considered types of consumer heterogeneity where the marginal distribution of the heterogeneity in the population is observed In BLP’s original paper, they include household income in the utility functions, and integrate out over the population income distribution (from the Current Population Survey) in simulating the predicted market shares November 17, 2016 se only 24 9:38 Econometric Models for Industrial Organization - 9in x 6in b2555-ch01 page 24 Econometric Models for Industrial Organization Another important example of this type of observed consumer heterogeneity is consumers’ location The idea is that the products are geographically differentiated, so that consumers might prefer choices which are located closer to their home Assume you want to model competition among movie theaters, as in Davis (2006) The utility of consumer i from theater j is: Uij = −αpj + β(Li − Lj ) + ξj + ij , where (Li −Lj ) denotes the geographic distance between the locations of consumer I and theater j The predicted market shares for each theater can be calculated by integrating out over the marginal empirical population density (i.e., integrating over the distribution of Li ) See also Thomadsen (2005) for a model of the fast-food industry, and Houde (2012) for retail gasoline markets The latter paper is noteworthy because instead of integrating over the marginal distribution of where people live, Houde integrates over the distribution of commuting routes He argues that consumers are probably more sensitive to a gasoline station’s location relative to their driving routes, rather than relative to their homes 1.7 Additional Details: General Presentation of Random Utility Models Introduce the social surplus function H(U ) ≡ E max (Uj + j∈J j) , where the expectation is taken over some joint distribution of ( , , J ) For each λ ∈ [0, 1], for all values of , and for any two vectors U and U , we have max(λUj + (1 − λ)Uj + j j) ≤ λ max(Uj + j j ) + (1 − λ) max(Uj j + j ) Since this holds for all vectors , it also holds in expectation, so that H(λU + (1 − λ)U ) ≤ λH(U ) + (1 − λ)H(U ) November 17, 2016 9:38 Econometric Models for Industrial Organization - 9in x 6in b2555-ch01 Demand Estimation for Differentiated-product Markets page 25 25 That is, H(·) is a convex function We consider its Fenchel–Legendre transformation2 defined as H ∗ (p) = max(p · U − H(U )), se only U where p is some J-dimensional vector of choice probabilities Because H is convex, we have that the FOCs characterizing H ∗ are p = ∇U H(U ) (1.6) Note that for discrete-choice models, this function is many-to-one For any constant k, H(U + k) = H(U ) + k, and hence if U satisfies p = ∇U H(U ), then also p = ∇U H(U + k) H ∗ (·) is also called the “conjugate” function of H(·) Furthermore, it turns out that the conjugate function of H ∗ (p) is just H(U ) — for this reason, the functions H ∗ and H have a dual relationship, and H(U ) = max(p · U − H ∗ (p)) p with, U ∈ ∂p H ∗ (p), (1.7) where ∂p H ∗ (p) denotes the subdifferential (or, synonymously, subgradiant or subderivative) of H ∗ at p For discrete choice models, this is typically a multivalued mapping (a correspondence) because ∇H(U ) is many-to-one.3 In the discrete choice literature, Eq (1.6) is called the William–Daly–Zachary theorem, and analogous to the Shepard/Hotelling lemmas, for the random utility model Equation (1.7) is a precise statement of the “inverse mapping” from choice probabilities to utilities for discrete choice models, and thus See Gelfand and Fomin (1965), Rockafellar (1971), Chiong, Galichon, and Shum (2013) Indeed, in the special case where ∇H(·) is one-to-one, then we have U = (∇H(p)) This is the case of the classical Legendre transform November 17, 2016 se only 26 9:38 Econometric Models for Industrial Organization - 9in x 6in b2555-ch01 Econometric Models for Industrial Organization reformulates (and is a more general statement of) the “inversion” result in Berry (1994) and Berry, Levinshon, and Pakes (1995) For specific assumptions on the joint distribution of (as with the generalized extreme value case cited previously), we can derive a closed form for the social surplus function H(U ), which immediately yield the choice probabilities via Eq (1.6) above For the MNL model, we know that K exp(Ui ) H(U ) = log i=0 From the conjugacy relation, we know that p = ∇H(U ) Normalizing U0 = 0, this leads to Ui = log(pi /p0 ) for i = 1, , K Plugging this back into the definition of H ∗ (p), we get that, K ∗ pi log(pi /p0 ) − log H (p) = i =0 K K pi (1.8) i =0 K pi log pi − log p0 = p0 i =1 pi + log p0 (1.9) i =1 K pi log pi = (1.10) i =0 To confirm, we again use the conjugacy relation U = ∇H ∗ (p) to get (for i = 0, 1, , K) that Ui = log pi Then imposing the normalization U0 = 0, we get that Ui = log(pi /p0 ) Bibliography Bajari, P and L Benkard (2005): “Demand Estimation With Heterogeneous Consumers and Unobserved Product Characteristics: A Hedonic Approach,” J Polit Econ., 113, 1239–1276 Berry, S (1994): “Estimating Discrete Choice Models of Product Differentiation,” RAND J Econ., 25, 242–262 Berry, S., J Levinsohn and A Pakes (1995): “Automobile Prices in Market Equilibrium,” Econometrica, 63, 841–890 page 26 November 17, 2016 9:38 Econometric Models for Industrial Organization - 9in x 6in b2555-ch01 se only Demand Estimation for Differentiated-product Markets page 27 27 Berry, S., J Levinsohn and A Pakes (1999): “Voluntary Export Restraints on Automobiles: Evaluating a Strategic Trade Policy,” Am Econ Rev., 89, 400–430 Bresnahan, T (1989): “Empirical Studies of Industries with Market Power,” in Handbook of Industrial Organization, eds by R Schmalensee and R Willig, vol North-Holland Chiong, K., A Galichon and M Shum (2013): “Duality in Dynamic Discrete Choice Models,” mimeo, Caltech Davis, P (2006): “Spatial Competition in Retail Markets: Movie Theaters,” RAND J Econ., 964–982 Gelfand, I and S Fomin (1965): Calculus of Variations Dover Goldberg, P (1995): “Product Differentiation and Oligopoly in International Markets: The Case of the US Automobile Industry,” Econometrica, 63, 891–951 Hausman, J (1996): “Valuation of New Goods under Perfect and Imperfect Competition,” in The Economics of New Goods, eds by T Bresnahan and R Gordon, pp 209–237 University of Chicago Press Houde, J (2012): “Spatial Differentiation and Vertical Mergers in Retail Markets for Gasoline,” Am Econ Rev., 102, 2147–2182 Keane, M (1994): “A Computationally Practical Simulation Estimator for Panel Data,” Econometrica, 62, 95–116 Maddala, G S (1983): Limited-dependent and Qualitative Variables in Econometrics Cambridge University Press McFadden, D (1978): “Modelling the Choice of Residential Location,” in Spatial Interaction Theory and Residential Location, ed by A K North Holland McFadden, D (1981): “Statistical Models for Discrete Panel Data,” in Econometric Models of Probabilistic Choice, eds C Manski and D McFadden MIT Press McFadden, D (1989): “A Method of Simulated Moments for Estimation of Discrete Response Models without Numerical Integration,” Econometrica, 57, 995–1026 Moon, R., M Shum and M Weidner (2012): “Estimation of Random Coefficients Logit Demand Models with Interactive Fixed Effects,” manuscript University of Southern California Nevo, A (2001): “Measuring Market Power in the Ready-to-eat Cereals Industry,” Econometrica, 69, 307–342 Petrin, A (2002): “Quantifying the Benefits of New Products: the Case of the Minivan,” J Polit Econ., 110, 705–729 Rockafellar, T (1971): Convex Analysis Princeton University Press Rosen, S (1974): “Hedonic Prices and Implicit Markets: Product Differentiation in Pure Competition,” J Polit Econ., 82, 34–55 November 17, 2016 28 9:38 Econometric Models for Industrial Organization - 9in x 6in b2555-ch01 Econometric Models for Industrial Organization se only Thomadsen, R (2005): “The Effect of Ownership Structure on Prices in Geographically Differentiated Industries,” RAND J Econ., 908–929 Trajtenberg, M (1989): “The Welfare Analysis of Product Innovations, with an Application to Computed Tomography Scanners,” J Polit Econ., 97(2), 444–479 page 28 November 28, 2016 13:58 Econometric Models for Industrial Organization - 9in x 6in b2555-index Index A Computational details for inner loop, 36–38 Computation of dynamic equilibrium, 76–77 Conditional choice probabilities (CCPs), 47 Conditional indirect utility functions, Conditional moment inequalities, 99 Confidence sets which cover identified parameters, 101–102 which cover the identified set, 102–105 Conjugate function, 25 Consumer choice, Continuation utility, 44 Controlled stochastic process, 56 Control variable, 29 Cost function parameters, estimating, 16–17 Cross-section of market shares, Crude frequency simulator, 59–60 Curse of dimensionality, 77 Affiliated PV models, 88–90 Aggregate market share, 5, 18 Auction models, 81–93 affiliated PV models, 88–90 affiliated values models, 88 Haile–Tamer’s incomplete model of English auctions, 92–93 nonparametric estimation, 85–88 parametric estimation, 81–85 testing between CV and PV, 90–92 B Backward recursion, 56 Bajari–Benkard–Levin (BBL) approach, 79 Bayesian posterior inference, 120 Behavioral model, 29–33 structural errors of model, 30 Bellman’s equation, 31, 37, 48, 57, 75, 80 Bertrand (price) competition, 15 BLP propose simulation methods, 18 Brute-force approach, 111 D Demand estimation, 2–3 applications, 22–24 Berry’s approach to estimate demand in differentiated product markets, 8–14 C Choice probabilities, 48 Choice-specific value function, 31, 48, 52 135 page 135 November 17, 2016 136 9:44 Econometric Models for Industrial Organization - 9in x 6in b2555-index Index demand analysis/estimate, reason, 1–3 for differentiated product markets, 1–26 discrete-choice approach to modeling demand, 4–8 measuring market power: recovering markups, 14–17 random utility models, 24–26 traditional approch to, 3–4 using random-coefficients logit model, 17–21 Deriving moment inequalities, 98 Discrete-choice models, 4, 25 Dynamic discrete-choice (DDC) models, 29 Markovian DDC models, 46 nonparametric identification of Markovian DDC models with unobserved state variables, 64–71 semiparametric identification of DDC models, 46–50 Dynamic equilibrium, computation of, 76–77 Dynamic games, 73–79 computation of dynamic equilibrium, 76–77 econometrics of dynamic oligopoly models, 73–74 games with incomplete information, 77–79 theoretical features, 74–76 Dynamic logit model, 35 Dynamic optimization models, 73 E Econometric model, 33–38 Econometrics of dynamic oligopoly models, 73–74 Eigenvalue-eigenvector decomposition, 70 Empirical distribution function (EDF), 86 Endogenous variable, Entry games with expectational errors, 99–100 with structural errors, 96–98 Equilibrium bidding strategy, 82 Estimating dynamic optimization models without numeric dynamic programming, 39 Euler’s constant, 42 Exclusion restriction, Exogenous variables, F Falsifiability: of model, 32 Fenchel–Legendre transformation, 25 Finite dimensional, 38 Finite-horizon DO problems, 56 Forward-looking model, 32 Forward-simulate value function, 46 Fudge factor, 117 G Games with incomplete information, 77–79 Geographic differentiation, 23 GHK simulator, 110–111 H Haile–Tamer’s incomplete model of English auctions, 92–93 Hedonic analysis, Hotz–Miller estimation scheme, 46 Hotz–Miller style conditional choice probabilities (CCP) approaches, 73 I Identified parameters, confidence sets which cover, 101–102 Implicit tax, 23 Importance sampling, 110 Incomplete information games, 96 Independence of Irrelevant Alternatives (IIA), Indirect approach, page 136 November 17, 2016 9:44 Econometric Models for Industrial Organization - 9in x 6in b2555-index Index Indirect utility function, Industrial organization (IO), Instrumental variable (IV) methods, Integrated value function, 52 Intuition, Inverse elasticity property, 2, 16 Inverse mapping, 25 IPV first-price auction model, nonparametrically identified, 87 Iteration procedure, 37 page 137 137 K Mileage transition probabilities, 35 MNL model, 6–8, 26 estimates for body CT scanners, 10 result for, 50–52 Modeling demand, discrete-choice approach to, Moment inequalities, 98 Monopoly, 1–2 Monopoly pricing model, Monte Carlo integration using the GHK simulator, 113–114 Multinomial logit (MNL) choice probabilities, Multinomial probit, Multiplayer version of Rust’s bus engine replacement model, 73 Myopic model, 32 Kernel density estimate of bid density, 86 N J Joint process, 58 J pricing first-order conditions, 15 L Logistic regression, 14 M Marginal cost, Marginal process, 58 Market power, Markov assumption, 75 Markov Chain Monte Carlo (MCMC) Simulation, 115–117 Markov chain theory, 116, 119 Markov dynamic choice models, 64–65 Markovian DDC models, 46 Markovian transition probabilities, 33 Markov-perfect equilibrium, 73 McFadden’s random utility framework see random utility framework McFadden’s social surplus function, 48 Metropolis–Hastings approach, 117–119 Nash equilibrium, 99 Nested estimation algorithm, 19 Nested estimation procedure, 20 Nested fixed point algorithm, 36 New empirical industrial organization (NEIO), Non-Gaussian Kalman filtering, 60 Nonlinear least squares (NLS), 11 Nonparametric estimation, 85–88 O Observed state variable, 30 Oligopoly case, Optimal stopping problem, 29, 56 P Parametric estimation, 81–85 Partial identification in structural models, 95–106 agnosticism, multiple equilibrium, and, 97–98 confidence sets which cover identified parameters, 101–102 confidence sets which cover the identified set, 102–105 November 17, 2016 138 9:44 Econometric Models for Industrial Organization - 9in x 6in b2555-index Index deriving moment inequalities, 98 entry games with expectational errors, 99–100 entry games with structural errors, 96–98 identified parameter vs identified set, 100–101 inference procedures with moment inequalities/ incomplete models, 100 random set approach, 105 Particle filtering, 60–64 Particle filter simulator, 64 Payoff-relevant state variables, 75 Perfect information game, 96 Predicted choice probabilities, 43 Predicted share function, 11 Price elasticity of demand, R Random coefficients, 17 Random coefficients logit model, 12 demand estimation using, 17–21 Random set approach, 105 sharp identified region for games with multiple equilibria, 106 Random utility framework, Random utility models, 48 Reduced form components, 47 Revenue Equivalence Theorem (RET), 83 Reversibility condition, 117 Roy’s Identity, Rust’s EV function, 52 S Semiparametric identification of DDC models, 46–50 Simulation methods, 109 Bayesian posterior inference, 120 GHK simulator, 110–111 importance sampling, 110 integrating over truncated (conditional) distribution, 114–115 Markov Chain Monte Carlo (MCMC) Simulation, 115–117 Metropolis–Hastings approach, 117–119 Monte Carlo integration using the GHK simulator, 113–114 Single-agent dynamic models, 29–71 behavioral model, 29–33 crude frequency simulator, 59–60 econometric model, 33–38 estimating dynamic optimization models without numeric dynamic programming, 39 flowchart for assumption, 66 likelihood function and simulation, 58–59 match hats to tildes, 43–46 model with persistence in unobservables, 55 nonparametric identification of Markovian dynamic discrete choice (DDC) models with unobserved state variables, 64–71 notation: hats and tildes, 40–43 Pakes patent renewal model, 55–58 particle filtering, 60–64 relations between different value function notions, 52–53 result for MNL model, 50–52 semiparametric identification of DDC models, 46–50 Slope coefficients, 17 Social surplus function, 24, 51, 53 Standard approach, Structural error(s), 5, 95 Supply-side (firm-side), IO theory focus on, page 138 November 17, 2016 9:44 Econometric Models for Industrial Organization - 9in x 6in b2555-index Index page 139 139 T V Transition probabilities, 38 Value function, 31 Voluntary export restraints (VERs), 23 U Unconditional inequalities, 100 Unobserved state variable, 30 Untruncated MVN density, 113 Utility maximization, Utility shock, W William–Daly–Zachary theorem, 25 ... 29 page 29 November 17, 2016 30 9:38 Econometric Models for Industrial Organization - 9in x 6in b2555-ch02 Econometric Models for Industrial Organization For simplicity, we describe the case... 17, 2016 36 9:38 Econometric Models for Industrial Organization - 9in x 6in b2555-ch02 Econometric Models for Industrial Organization Defining u ¯(x, i; θ) ≡ u(x, , i; θ) − the form Prob (it |xt... Shum (2013) for the most general treatment of this November 17, 2016 38 9:38 Econometric Models for Industrial Organization - 9in x 6in b2555-ch02 Econometric Models for Industrial Organization