Workshops

12月15日（金 Friday） 4:50-6:10

東京大学大学院経済学研究科棟 新棟3階 第3教室
at theLecture Hall No.3 on the 3rd floor of the New Economics Building

Yoon-Jae Whang (Seoul National University)

Testing for Non-nested Conditional Moment Restrictions via Conditional Empirical Likelihood [PDF]

We propose non-nested tests for competing conditional moment restriction models using a method of conditional empirical likelihood, recently suggested by Kitamura, Tripathi and Ahn (2004) and Zhang and Gijbels (2003). We use the implied conditional probabilities to define our test statistics, which take into account the full implications of conditional moment restrictions. We develop three types of non-nested tests: the moment encompassing, Cox-type, and efficient score encompassing tests. We derive the asymptotic null distributions and investigate their power properties against a sequence of local alternatives and a fixed global alternative. Our tests have power proerties that are very distinct from some of the existing tests based on finite-dimensional unconditional moment restrictions and are consistent against alternatives that cannot be detected by the latter type tests. In particular, if the support of the moment function is bounded, our Cox-type test is consistent against all departures from the null hypothesis toward the non-nested alternative hypothesis under very mild conditions. On the other hand, the moment encompassing and efficient score encompassing tests require some additional assumptions for consistency which guarantee the non-centrality parameters to be non-zero. Simulation experiments show that our tests have reasonable finite sample properties.

4月14日（金 Friday） 4:50-6:10

東京大学大学院経済学研究科棟 新棟3階 第3教室
at theLecture Hall No.3 on the 3rd floor of the New Economics Building

本田敏雄  T. Honda （一橋大学経済学研究科）
Nonparametric estimation for linear processes

The talk will be based on two discussion papers :
1."Nonparametric least absolute deviation regression for
long-range dependent processes,"
2. "Nonparametric density estimation for linear porcesses with
infinite variance,"
which are available at http://www.econ.hit-u.ac.jp/~kenkyu/ jpn/pub/dp-top.htm.
The former deals with estimation of the conditional median of Y_t w.r.t. X_t when both Y_t and X_t show long-range dependence.  The latter is about the asymptotic properties of the kernel density estimator when we observe a linear process with heavy-tailed i.i.d. innovations. The asymptotic properties depend on the coefficients of the linear process and the tail condition of the innovations.

4月28日（金 Friday） 4:50-6:10

東京大学大学院経済学研究科棟 新棟3階 第3教室
at theLecture Hall No.3 on the 3rd floor of the New Economics Building

塚原英敦  Hideatsu Tsukahara （成城大学＆ETH）

One-parameter Families of distortion Risk Measures

*統計数理研究所・リスク解析センタープロジェクトと共催

This paper introduces parametric families of distortion risk measures, investigates their properties, and discusses its use in risk management. Their derivation is based on Kusuoka’s representation theorem of law invariant and comonotonically additive coherent risk measures.  Much simplified proofs of some known results under slightly weaker condition are also given. Our approach is then to narrow down a tractable class of risk measures by requiring their comparability with the traditional expected shortfall. We make numerical comparison among them, and propose a method of estimating the value of the distortion risk measures based on data, Their use and interpretation in risk management will also be discussed.

5月12日（金 Friday） 4:50-6:10

東京大学大学院経済学研究科棟 新棟3階 第3教室
at the Lecture Hall No.3 on the 3rd floor of the New Economics Building

山下智志  S. Yamashita　（統計数理研究所）

債権回収率計量化モデルの系譜

*統計数理研究所・リスク解析センタープロジェクトと共催

信用リスクを構成する要因のうち、デフォルト確率(PD)については多くのモデルが存在し、実用化されつつある。しかし、デフォルトした後の債券の回収率もしくはデフォルト時損失率（LGD）については、未だ方法論が確立せず、実用的なモデルは開発途上である。今回はこれまで学界や実業界で提案されている債券回収率の数理モデルを紹介し、それぞれのモデルや方法がもつ本源的アイデアと、長所・短所ついて議論する。さらに実務的な債券回収率モデルを構築する際、解決すべき課題と今後の方向性について検討する。

5月26日（金 Friday） 4:50-6:10

東京大学大学院経済学研究科棟 新棟3階 第3教室
at the Lecture Hall No.3 on the 3rd floor of the New Economics Building

S. Chen （Hong Kong University of Science and Technology）

Nonparametric Identification and Estimation of Truncated Regression Models

We provide a comprehensive treatment of nonparametric identification of truncated regression models for both the cross-sectional and panel data settings.  We also propose non-parametric estimation procedures based on idfentification results.  Our results are robust to misspecification of parametric structures of the models.

連続講義シリーズ「保険と金融のための応用確率論」

5月16日（火）、19日（金）、23日（火）、30日（火） 5:00-6:30

東京大学大学院経済学研究科棟 新棟12階 第3共同研究室

at the Conference Room No.3 on the 12th floor of the New Economics Building

西山陽一　Y. Nishiyama （統計数理研究所）

確率過程の統計学：マルチンゲールとempirical process

*CIRJEプロジェクト「保険と金融の統計学」主催

*統計数理研究所・リスク解析センタープロジェクト共催

CIRJEプロジェクト「保険と金融の統計学」では統計数理研究所のリスク解析センターと 共同して、統計的金融リスク管理など近年の金融・保険リスクの統計的分析の 確率論・統計学の数理的基礎について、世界的に活躍中の若手研究者による大学院生向けの特別な連続講義を計画しています。 関心のある大学院生の参加を歓迎しますが、 受講者には経済学部等で扱われる基礎的な確率論と数理統計学の基本的知識を仮定します。

6月16日（金 Friday） 4:50-6:10

東京大学大学院経済学研究科棟 新棟3階 第3教室
at the Lecture Hall No.3 on the 3rd floor of the New Economics Building

Yuichi Kitamura（Yale University）

Empirical Likelihood Methods in Econometrics: Theory and Practice

Recent developments in empirical likelihood (EL) are reviewed.First, to put the method in perspective, two interpretations of empirical likelihood are presented, one as a nonparametric maximum likelihood estimation method (NPMLE) and the other as a generalized minimum contrast estimator (GMC). The latter interpretation provides a clear connection between EL, GMM, GEL and other related estimators. Second, EL is shown to have various advantages over other methods. The theory of large deviations demonstrates that EL emerges naturally in achieving asymptotic optimality both for estimation and testing. Interestingly, higher order asymptotic analysis also suggests that EL is generally a preferred method. Third, extensions of EL are discussed in various settings, including estimation of conditional moment restriction models, nonparametric specification testing and time series models. Finally, practical issues in applying EL to real data, such as computational algorithms for EL, are discussed.  Numerical examples to illustrate the efficacy of the method are presented.

6月23日（金 Friday） 4:50-6:10

東京大学大学院経済学研究科棟 新棟3階 第3教室
at the Lecture Hall No.3 on the 3rd floor of the New Economics Building

Masayuki Hirukawa （Concordia University）

HAC Estimation: Bandwidth Choice and GMM Bootstrap

Two topics on HAC estimation are presented.  The first half discusses the bandwidth choice.  From the viewpoint of a parallel setting to probability density estimation, the theory of two-stage plug-in bandwidth selection and a reliable implementation method are developed.  The second half considers an improvement in the GMM bootstrap for linear time series regression.  The GMM boostrap by Inoue and Shintani (2006) requires a higher-order kernel for the HAC estimation in order to attain the asymptotic refinements.  Employing the inverse of the NPW-HAC estimator (Xiao and Linton, 2002; Hirukawa, 2006) as the weighting matrix can stabilize the GMM bootstrap and thus improve its finite-sample properties.

連続講義シリーズ「保険と金融のための応用確率論」

6月13日（火）、20日（火）、27日（火）、7月4日（火） 5:00-6:30

東京大学大学院経済学研究科棟 新棟13階 第3共同研究室

at the Conference Room No.3 on the 13th floor of the New Economics Building

志村隆彰　T. Shimura （統計数理研究所）

Regular Variationの理論と確率論

*CIRJEプロジェクト「保険と金融の統計学」主催

*統計数理研究所・リスク解析センタープロジェクト共催

CIRJEプロジェクト「保険と金融の統計学」では統計数理研究所のリスク解析センターと 共同して、統計的金融リスク管理など近年の金融・保険リスクの統計的分析の 確率論・統計学の数理的基礎について、世界的に活躍中の若手研究者による大学院生向けの特別な連続講義を計画しています。 関心のある大学院生の参加を歓迎しますが、 受講者には経済学部等で扱われる基礎的な確率論と数理統計学の基本的知識を仮定します。

7月7日（金 Friday） 4:50-6:10

東京大学大学院経済学研究科棟 新棟3階 第3教室
at the Lecture Hall No.3 on the 3rd floor of the New Economics Building

Taisuke Otsu （Yale University）

Asymptotic Optimality of Empirical Likelihood for Selecting Moment Restrictions

This paper studies large deviation optimal properties of the empirical likelihood sequential testing (ELST) procedures for selecting moment restrictions. Since moment selection problems have discrete parameter spaces, the Pitman-type local alternative approach is not very helpful. By the theory of large deviations, we analyze convergence rates of error probabilities under fixed distributions. We propose three optimal properties of the ELST procedures: (i) the generalized Neyman-Pearson optimality, (ii)
the overestimation error optimality, and (iii) the minimax misclassification error optimality.

7月14日（金 Friday） 4:50-6:10

東京大学大学院経済学研究科棟 新棟3階 第3教室
at the Lecture Hall No.3 on the 3rd floor of the New Economics Building

Kouske Imai （Princeton University）

Statistical Analysis of Randomized Experiments with Nonignorable Missing Outcomes

Missing data are frequently encountered in the statistical analysis of randomized experiments. In this paper, I study the problem of nonignorable missing outcomes where the missing-data mechanism of an outcome variable may depend on the values of the variable itself. I first consider a standard randomized experiment, and propose the maximum likelihood and Bayes estimators which model nonignorable missing out-comes I then generalize the proposed methodology to randomized experiments with noncompliance, and compare it with the latent ignorability approach commonly used in the literature. Monte Carlo experiments are conducted to investigate the finite-sample properties of the proposed estimators. When sample size is small, the Bayes estimator is shown to outperform the maximum likelihood estimator. The relative advantage of the Bayes estimator is large when noncompliance exists. Finally, I apply the pro-posed method to analyze data from a German election experiment and an influenza vaccination study.

7月21日（金 Friday） 4:50-6:10

東京大学大学院経済学研究科棟 新棟3階 第3教室
at the Lecture Hall No.3 on the 3rd floor of the New Economics Building

Ed Vytlacil （Columbia University）

Threshold Crossing Models and Bounds on Treatment Effects:  A Nonparametric Analysis

This paper considers the evaluation of the average treatment effect of a binary endogenous regressor on a binary outcome when one imposes a threshold crossing model on both the endogenous regressor and the outcome variable but without imposing parametric functional form or distributional assumptions. Without parametric restrictions, the average effect of the binary endogenous variable is not generally point identified. This paper constructs sharp bounds on the average effect of the endogenous variable that exploit the structure of the threshold crossing models and any exclusion restrictions. We also develop methods for inference on the resulting bounds.

7月28日（金 Friday） 11:00-12:00 *通常と時間が違いますのでご注意下さい

東京大学大学院経済学研究科棟 新棟3階 第3教室
at the Lecture Hall No.3 on the 3rd floor of the New Economics Building

Sokbae Lee （University College London）

Nonparametric Instrumental Variables Estimation of a Quantile Regression Model

We consider nonparametric estimation of a regression function that is identified by requiring a specified quantile of the regression error conditional on an instrumental variables to be zero.  The resulting estimating equation is a nonlinear integral equation of
the first kind, which generates an ill-posed-inverse problem.

8月16日（水 Wednesday） 16:00-17:30

東京大学大学院経済学研究科棟 新棟3階 第3教室
at theLecture Hall No.3 on the 3rd floor of the New Economics Building

↑以前と変更になりましたのでご注意下さい。

Yacine Ait-Sahalia （Princeton University）

Portfolio Choice with Jumps:  A Closed Form Solution

＊東京ファイナンス研究会のワークショップです。詳しくはこちらをご覧下さい→東京大学金融教育研究センター（CARF）のホームページ

9月19日（火 Tuesday） 15:00-16:40

応用統計輪講と共催

東京大学大学院経済学研究科棟 新棟3階 第2教室
at theLecture Hall No.2 on the 3rd floor of the New Economics Building

Andrew Harvey (Cambridge University）

Time-Varying Quantiles

A time-varying quantile can be fitted to a sequence of observations by formulating a time series model for the corresponding population quantile and iteratively applying a suitably modified state space signal extraction algorithm. Quantiles estimated in this way provide information on various aspects of a time series, including dispersion, asymmetry and, for financial applications, value at risk. Tests for the constancy of quantiles, and associated contrasts, are constructed using indicator variables; these tests have a similar form to stationarity tests and, under the null hypothesis, their asymptotic distributions belong to the Cram?r von Mises family. Estimates of the quantiles at the end of the series provide the basis for forecasting. As such they offer an alternative to conditional quantile autoregressions and, at the same time, give some insight into their structure and potential drawbacks.

9月22日（金 Friday） Part I: 10:00-11:00, Part II: 11:10-12:10

*COE Frontier Economics Lecture Seriesと共催

東京大学大学院経済学研究科棟 新棟3階 第2教室
at theLecture Hall No.2 on the 3rd floor of the New Economics Building

Costas Meghir (University College of London）

A Recent Progress on the Measurement of the Returns to Formal and Informal Training on the Life Cycle Earnings [paper]

9月22日（金 Friday） Part I: 14:00-15:00, Part II: 15:10-16:10

*COE Frontier Economics Lecture Series と共催

東京大学大学院経済学研究科棟 新棟3階 第2教室
at theLecture Hall No.2 on the 3rd floor of the New Economics Building

Jean-Marc Robin (University of Paris I and University College of London）

Noisy Independent Component Analysis with Application to Econometrics

[paper1]  [slides1]   [paper2]  [slides2]

11月10日（金 Friday） 4:50-6:10

東京大学大学院経済学研究科棟 新棟3階 第3教室
at theLecture Hall No.3 on the 3rd floor of the New Economics Building

Patrik Guggenberger （UCLA)

Subsampling methods in Econometrics

Papers are downloadable from the following website:

http://www.econ.ucla.edu/people/defaultpapers.cfm?NAME=Guggenberger

(1) "The limit of finite sample size and a problem with subsampling"

This paper considers inference based on a test statistic that has a limit distribution that is discontinuous in a nuisance parameter or the parameter of interest. The paper shows that subsample, b_n<n bootstrap, and standard fixed critical value tests based on such a test statistic often have asymptotic size - defined as the limit of the finite sample size - that is greater than the nominal level of the tests. We determine precisely the asymptotic size of such tests under a general set of high-level conditions that are relatively easy to verify. The high-level conditions are verified in several examples. Analogous results are established for confidence intervals. The results apply to tests and confidence intervals (i) for parameters that may be near a boundary, (ii) for parameters defined by moment inequalities, (iii) based on super-efficient or shrinkage estimators, (iv) based on post-model selection estimators, (v) in autoregressive models with roots that may be close to unity, (vi) in predictive regression models with nearly-integrated regressors, (vii) in models with lack of identification at some point(s) in the parameter space, such as models with weak instruments and threshold autoregressive models, (viii) for non-differentiable functions of parameters, and (ix) for differentiable functions of parameters that have zero first-order derivative.

(2) " Hybrid and size-corrected subsample methods"

TBA

11月14日（火 Tuesday） 3:00-4:40

統計輪講と共催

東京大学大学院経済学研究科棟 新棟3階 第3教室
at theLecture Hall No.3 on the 3rd floor of the New Economics Building

John Geweke（University of Iowa）

Smoothly Mixing Regressions [PDF]

This paper extends the conventional Bayesian mixture of normals model by permitting state probabilities to depend on observed covariates. The dependence is captured by a simple multinomial probit model. A conventional and rapidly mixing MCMC algorithm provides access to the posterior distribution at modest computational cost. This model is
competitive with existing econometric models, as documented in the paper's illustrations. The first illustration studies quantiles of the distribution of earnings of men conditional on age and education, and shows that smoothly mixing regressions are an attractive alternative to non-Baeysian quantile regression. The second illustration models serial
dependence in the S&P 500 return, and shows that the model compares favorably with ARCH models using out of sample likelihood criteria.

11月13日（月 Monday）-11月15日（水 Wednesday）

東京大学山上会館大会議室
at Sanjo Conference Hall of University of Tokyo (Hongo Campus)
(Campus Map)  http://www.u-tokyo.ac.jp/index_j.html

国際研究集会　[Program]

International Symposium on Statistical Analysis of Spatio-Temporal Data

Sponsors:　科学研究費基盤研究(A)「時空間統計解析の理論と応用」
　　　　　　 (課題番号：15200021研究期間：平成１５年度〜平成１８年度)
　　　　　　21世紀COEプログラム「 市場経済と非市場機構との連関研究拠点 」
　　　　　　東京大学大学院経済学研究科

※　日本統計学会75周年記念事業

Invited Speakers:
                   N.A.C Cressie (Ohio State University)
                   W.T.M. Dunsmuir (University of New South Wales)
                   L. Gobillon (Institut National d'Etudes Demographiques)
                   V. De Oliveira (University of Texas)
                   P.M. Robinson (London School of Economics)
                   M.L. Stein (University of Chicago)
                   D. Tjostheim (University of Bergen)
                   J. Zidek (University of British Columbia

12月1日（金 Friday） 4:50-6:10

東京大学大学院経済学研究科棟 新棟3階 第3教室
at theLecture Hall No.3 on the 3rd floor of the New Economics Building

Tore Schweder（University of Oslo, Department of   Economics）

Distributional inference by confidence curves [paper] [handout]

Please refer to the relevant websites:

Schweder and Hjort (2002)

Schweder (2003)

Fisher (1930) proposed the fiducial distribution as an alternative to the Bayesian posterior. Neyman (1941) showed that confidence intervals are obtained from the fiducial quantiles in nice models. Fisher and Neyman could not agree, and fiducial inference fell into disrespect. The confidence curve of Birnbaum (1961) displays confidence intervals for all degrees of confidence, and is based on the cumulative confidence (or fiducial) distribution. Confidence curves, also for vector parameters, are obtained by mapping the (profile) deviance function into the unit interval by the cdf of the deviance at the true value. I will briefly present this basic theory of frequentist distributional inference, and discuss issues of bias, power, and optimality. I will also present product confidence curves representing nested families of simultaneous confidence bands for high dimensional parameters. They are obtained from bootstrapping by the abc-method. I will apply the methods to Photo-ID data on bowhead whales, and to Norwegian personal income data.

12月8日（金 Friday） 4:50-6:10

東京大学大学院経済学研究科棟 新棟3階 第3教室
at theLecture Hall No.3 on the 3rd floor of the New Economics Building

Azeem M. Shaikh（Postdoctoral Fellow at Yale University, Assistant Professor at University of Chicago）

Inference for Partially Identified Models [paper1] [paper2]

(paper1)``Inference for the Identified Set in Partially Identified Econometric Models"
This paper provides computationally intensive, yet feasible methods for inference in a very general class of partially identified econometric models. Let P denote the distribution of the observed data. The class of models we consider are defined by a population objective function Q(θ, P) for θ∈Θ. The point of departure from the classical extremum estimation framework is that it is not assumed that Q(θ, P) has a unique minimizer in the parameter space Θ. The goal may be either to draw inferences about some unknown point in the set of  minimizers of the population objective function or to draw inferences about the set of minimizers itself. In this paper, the object of interest is Θ0(P) = argminθ∈Θ Q(θ,P), and so we seek random sets that contain this set with at least some prespecified probability asymptotically. We also consider situations where the object of interest is the image ofΘ0(P) under a known function. Random sets satisfying the desired coverage property are constructed under weak assumptions. Conditions are provided under which the confidence regions are asymptotically valid not only pointwise in P, but also uniformly in P. Finally, we illustrate the use of our methods with an empirical study of the impact of top-coding of outcomes on inferences about the parameters of a linear regression.

(paper2)``Inference for Identifiable Parameters in Partially Identified Econometric Models"

This paper considers the problem of inference for partially identified econometric models. The class of models studied are defined by a population objective function Q(θ, P) for θ∈ Θ. The second argument indicates the dependence of the objective function on P, the distribution of the observed data. Unlike the classical extremum estimation framework, it is not assumed that Q(θ, P) has a unique minimizer in the parameter space
Θ. The goal may be either to draw inferences about some unknown point in the set of minimizers of the population objective function or to draw inferences about the set of minimizers itself. In this paper, the object of interest is some unknown point θ ∈ Θ0(P), where Θ0(P) = argminθ∈Θ Q(θ, P), and so we seek random sets that contain each θ∈ Θ0(P) with at least some prespecified probability asymptotically. We also consider situations where the object of interest is the image of some point θ∈ Θ0(P) under a known function. Computationally intensive, yet feasible procedures for constructing random sets satisfying the desired coverage property under weak assumptions are provided. We also provide conditions under which the confidence regions are uniformly consistent in level. To do this, we first derive new uniformity results about subsampling that are of independent interest.