京都大学 / 0048 / 新制・課程博士 / 博士(経済学) / 甲第19164号 / 経博第518号 / 新制||経||274(附属図書館) / 32156 / 京都大学大学院経済学研究科経済学専攻 / (主査)教授 西山 慶彦, 准教授 奥井 亮, 准教授 山田 憲 / 学位規則第4条第1項該当 / Doctor of Economics / Kyoto University / DFAM

<p>This dissertation consists of three chapters relating to</p> <p>identification and inference in dynamic microeconometric models</p> <p>including dynamic discrete games with many players, dynamic games with</p> <p>discrete and continuous choices, and semiparametric binary choice and</p> <p>duration panel data models.</p> <p>The first chapter provides a framework for estimating large-scale</p> <p>dynamic discrete choice models (both single- and multi-agent models)</p> <p>in continuous time. The advantage of working in continuous time is</p> <p>that state changes occur sequentially, rather than simultaneously,</p> <p>avoiding a substantial curse of dimensionality that arises in</p> <p>multi-agent settings. Eliminating this computational bottleneck is</p> <p>the key to providing a seamless link between estimating the model and</p> <p>performing post-estimation counterfactuals. While recently developed</p> <p>two-step estimation techniques have made it possible to estimate</p> <p>large-scale problems, solving for equilibria remains computationally</p> <p>challenging. In many cases, the models that applied researchers</p> <p>estimate do not match the models that are then used to perform</p> <p>counterfactuals. By modeling decisions in continuous time, we are able</p> <p>to take advantage of the recent advances in estimation while</p> <p>preserving a tight link between estimation and policy experiments. We</p> <p>also consider estimation in situations with imperfectly sampled data,</p> <p>such as when we do not observe the decision not to move, or when data</p> <p>is aggregated over time, such as when only discrete-time data are</p> <p>available at regularly spaced intervals. We illustrate the power of</p> <p>our framework using several large-scale Monte Carlo experiments.</p> <p>The second chapter considers semiparametric panel data binary choice</p> <p>and duration models with fixed effects. Such models are point</p> <p>identified when at least one regressor has full support on the real</p> <p>line. It is common in practice, however, to have only discrete or</p> <p>continuous, but possibly bounded, regressors. We focus on</p> <p>identification, estimation, and inference for the identified set in</p> <p>such cases, when the parameters of interest may only be partially</p> <p>identified. We develop a set of general results for</p> <p>criterion-function-based estimation and inference in partially</p> <p>identified models which can be applied to both regular and irregular</p> <p>models. We apply our general results first to a fixed effects binary</p> <p>choice panel data model where we obtain a sharp characterization of</p> <p>the identified set and propose a consistent set estimator,</p> <p>establishing its rate of convergence under different conditions.</p> <p>Rates arbitrarily close to <italic>n<super>-1/3</super></italic> are</p> <p>possible when a continuous, but possibly bounded, regressor is</p> <p>present. When all regressors are discrete the estimates converge</p> <p>arbitrarily fast to the identified set. We also propose a</p> <p>subsampling-based procedure for constructing confidence regions in the</p> <p>models we consider. Finally, we carry out a series of Monte Carlo</p> <p>experiments to illustrate and evaluate the proposed procedures. We</p> <p>also consider extensions to other fixed effects panel data models such</p> <p>as binary choice models with lagged dependent variables and duration</p> <p>models.</p> <p>The third chapter considers nonparametric identification of dynamic</p> <p>games of incomplete information in which players make both discrete</p> <p>and continuous choices. Such models are commonly used in applied work</p> <p>in industrial organization where, for example, firms make discrete</p> <p>entry and exit decisions followed by continuous investment decisions.</p> <p>We first review existing identification results for single agent</p> <p>dynamic discrete choice models before turning to single-agent models</p> <p>with an additional continuous choice variable and finally to</p> <p>multi-agent models with both discrete and continuous choices. We</p> <p>provide conditions for nonparametric identification of the utility</p> <p>function in both cases.</p> / Dissertation

Industrirobotar är idag en väsentlig del i tillverkningsindustrin där de bland annat används för att minska kostnader, öka produktivitet och kvalitet och ersätta människor i farliga eller slitsamma uppgifter. Höga krav på noggrannhet och snabbhet hos robotens rörelser innebär också höga krav på de matematiska modeller som ligger till grund för robotens styrsystem. Modellerna används där för att beskriva det komplicerade sambandet mellan robotarmens rörelser och de motorer som orsakar rörelsen. Tillförlitliga modeller är också nödvändiga för exempelvis mekanisk design, simulering av prestanda, diagnos och övervakning. En trend idag är att bygga lättviktsrobotar, vilket innebär att robotens vikt minskas men att den fortfarande kan hantera en lika tung last. Orsaken till detta är främst att minska kostnaden, men också säkerhetsaspekter spelar in. En lättare robotarm ger dock en vekare struktur där elastiska effekter inte längre kan försummas i modellen om man kräver hög prestanda. De elastiska effekterna beskrivs i den matematiska modellen med hjälp av fjädrar och dämpare. Denna avhandling handlar om hur dessa matematiska modeller kan tas fram genom systemidentifiering, vilket är ett viktigt verktyg där mätningar från robotens rörelser används för att bestämma okända parametrar i modellen. Det som mäts är position och moment hos robotens alla motorer. Identifiering av industrirobotar är ett utmanande problem bland annat eftersom robotens beteende varierar beroende på armens position. Den metod som föreslås i avhandlingen innebär att man först identifierar lokala modeller i ett antal positioner. Var och en av dessa beskriver robotens beteende kring en viss arbetspunkt. Sedan anpassas parametrarna i en global modell, som är giltig för alla positioner, så att den så väl som möjligt beskriver det lokala beteendet i de olika positionerna. I avhandlingen analyseras olika metoder för att ta fram lokala modeller. För att få bra resultat krävs att experimenten är omsorgsfullt utformade. För att minska osäkerheten i den globala modellens identifierade parametrar ingår också valet av optimala positioner för experimenten. Olika metoder för att identifiera parametrarna jämförs i avhandlingen och experimentella resultat visar användbarheten av den föreslagna metoden. Den identifierade robotmodellen ger en bra global beskrivning av robotens beteende. Resultatet av forskningen har även gjorts tillgängligt i ett datorverktyg för att noggrant kunna ta fram lokala modeller och identifiera parametrar i dynamiska robotmodeller. / Industrial robots are today essential components in the manufacturing industry where they are used to save costs, increase productivity and quality, and eliminate dangerous and laborious work. High demands on accuracy and speed of the robot motion require that the mathematical models, used in the motion control system, are accurate. The models are used to describe the complicated nonlinear relation between the robot motion and the motors that cause the motion. Accurate dynamic robot models are needed in many areas, such as mechanical design, performance simulation, control, diagnosis, and supervision. A trend in industrial robots is toward lightweight robot structures, where the weight is reduced but with a preserved payload capacity. This is motivated by cost reduction as well as safety issues, but results in a weaker (more compliant) mechanical structure with enhanced elastic effects. For high performance, it is therefore necessary to have models describing these elastic effects. This thesis deals with identification of dynamic robot models, which means that measurements from the robot motion are used to estimate unknown parameters in the models. The measured signals are angular position and torque of the motors. Identifying robot models is a challenging task since an industrial robot is a multivariable, nonlinear, unstable, and resonant system. In this thesis, the unknown parameters (typically spring-damper pairs) in a physically parameterized nonlinear dynamic model are identified, mainly in the frequency domain, using estimates of the nonparametric frequency response function (FRF) in different robot configurations/positions. Each nonparametric FRF then describe the local behavior around an operating point. The nonlinear parametric robot model is linearized in the same operating points and the optimal parameters are obtained by minimizing the discrepancy between the nonparametric FRFs and the parametric FRFs (the FRFs of the linearized parametric robot model). Methods for estimating the nonparametric FRF from experimental data are analyzed with respect to bias, variance, and nonlinearities. In order to accurately estimate the nonparametric FRF, the experiments must be carefully designed. To minimize the uncertainty in the estimated parameters, the selection of optimal robot configurations/positions for the experiments is also part of the design. Different parameter estimators are compared in the thesis and experimental results show the usefulness of the proposed identification procedure. The identified nonlinear robot model gives a good global description of the dynamics in the frequency range of interest. The research work is also implemented and made easily available in a software tool for accurate estimation of nonparametric FRFs as well as parametric robot models.