此篇文章探討，使用離散選擇模型(discrete choice model)中的邏輯模型(logit model)分析，若資料具有遺漏值(incomplete-data)，比較將具有遺漏值樣本值皆移除與使用多重插補方法補值之參數估計結果。 本文使用的多重差補法為Buuren(2007)等人所提出的Multiple Imputation by Chained Equation(MICE)多重插補方法進行補值，並使用Rubin(1987)所提出的方法合併參數估計結果。從模擬結果之參數偏誤盒狀圖可知插補後參數估計與設定參數差異不大，另外插補次數對於參數估計結果影響不大，且在遺漏比例(missing percentage)大時，參數估計結果比起將具有遺漏值樣本直接移除的參數估計較為穩定。 另外使用實際資料分析，發現具有遺漏值樣本直接移除的參數估標準差比起插補後參數估計標準差大的趨勢，與模擬結果相同。 / This paper focuses on using discrete choice logit model to analyze incompleted data. To deal with the incompleted data, complete case analysis and multiple imputation are used, and compare the result of parameter estimates of the two methods. The method of multiple imputation which this paper used is Multiple Imputation by Chained Equation (MICE). With the estimates from multiple imputed data sets, using Rubin’s method (1987) to pool the estimates. The simulation shows that after imputing the missing values, the estimates from the imputed data are not much difference from the real parameters. The number of imputation does not effect the estimates much. With larger missing percentage, the estimates from the imputed data is more robust than the estimates from the complete case analysis. In real data analysis, the standard deviation of estimates from using complete case analysis are bigger than imputed data, this result is the same with the simulation.

In recent decades, economists are gradually showing their interests in the study of happiness. They even put forward some challenges to the traditional theories. In contrast, studies on Chinese happiness problem are not enough in terms of breadth and depth.  This paper used the data provided by China General Social Survey to conduct an empirical analysis. The model author adopted is Ordered Discrete Choice model. In the empirical section, author analyzed the impact of income, macroeconomic variables, etc.  Ultimately, based on the empirical results, author proposed some policy recommendations and further study suggestions.

[pt] Este artigo investiga o impacto da provisão pública na concorrência dentro do setor de ensino superior brasileiro. Desenvolvemos e estimamos um modelo empírico de demanda por ensino superior que incorpora instituições gratuitas e restrições de escolha do consumidor. Nosso modelo produz padrões de substituição mais realistas do que um modelo logit sem restrições, indicando que a seletividade é uma dimensão essencial da diferenciação de produtos. Mostramos que os programas públicos mais seletivos exercem pressão competitiva comparável aos programas privados mais seletivos, mas os programas públicos menos seletivos exercem mais pressão competitiva do que os programas privados menos seletivos. Nossas estimativas da resposta de oferta das instituições privadas sugerem que, na ausência de programas públicos, as mensalidades seriam cerca de 7 por cento mais altas. Esses resultados fornecem informações importantes sobre a dinâmica competitiva do ensino superior brasileiro e destacam o papel da provisão pública na promoção da concorrência nesse setor. / [en] This paper investigates the impact of public provision on competition within the Brazilian higher education sector. We develop and estimate an empirical model of demand for higher education that incorporates tuition-free institutions and consumer choice constraints. Our model produces more realistic substitution patterns than a logit model without constraints, indicating that selectivity is an essential dimension of product differentiation. We find that the most selective public programs exert comparable competitive pressure to the most selective private programs, but the least selective public programs exert more competitive pressure than the least selective private programs. Our estimates of the supply response of private institutions suggest that, in the absence of public programs, tuitions would be about 7 percent higher. These findings provide important insights into the competitive dynamics of Brazilian higher education and highlight the role of public provision in promoting competition in this sector.

No description available.

This study proposes several decision-making tools utilizing optimization and machine learning frameworks to assess and improve the safety of the workplaces. The first chapter of this study presents a novel mathematical model to optimally locate a set of detectors to minimize the expected number of casualties in a given threat area. The problem is formulated as a nonlinear binary integer programming model and then solved as a linearized branch-and-bound algorithm. Several sensitivity analyses illustrate the model's robustness and draw key managerial insights. One of the prevailing threats in the last decades, Active Shooting (AS) violence, poses a serious threat to public safety. The second chapter proposes an innovative mathematical model which captures several essential features (e.g., the capacity of the facility and individual choices, heterogeneity of individual behavioral and choice sets, restriction on choice sets depending on the location of the shooter and facility orientation, and many others) which are essential for appropriately characterizing and analyzing the response strategy for civilians under an AS exposed environment. We demonstrate the applicability of the proposed model by implementing the effectiveness of the RUN.HIDE.FIGHT.® (RHF) program in an academic environment. Given most of the past incidents took place in built environments (e.g., educational and commercial buildings), there is an urgent need to methodologically assess the safety of the buildings under an active shooter situation. Finally, the third chapter aims to bridge this knowledge gap by developing a learning technique that can be used to model the behavior of the shooter and the trapped civilians in an active shooter incident. Understanding how the civilians responded to different simulated environments, a number of actions could have been undertaken to bolster the safety measures of a given facility. Finally, this study provides a customized decision-making tool that adopts a tailored maximum entropy inverse reinforcement learning algorithm and utilizes safety measurement metrics, such as the percentage of civilians who can hide/exit in/from the system, to assess a workplace's safety under an active shooter incident.

京都大学 / 新制・課程博士 / 博士(農学) / 甲第24681号 / 農博第2564号 / 新制||農||1100(附属図書館) / 学位論文||R5||N5462(農学部図書室) / 京都大学大学院農学研究科生物資源経済学専攻 / (主査)教授 栗山 浩一, 教授 浅見 淳之, 教授 伊藤 順一 / 学位規則第4条第1項該当 / Doctor of Agricultural Science / Kyoto University / DGAM

Die vorliegende Studienarbeit beschäftigt sich mit der Fragestellung, inwiefern sich Revealed-Preference-Daten aus der deutschen Mobilitätsbefragung "Mobilität in Städten" SrV 2008 dazu eignen, um basierend auf denen im Datensatz enthaltenen Wegen Verkehrsmittelwahlmodelle zu schätzen. Dazu wurden Wegedaten aus der Befragung verwendet, und die Wahlalternativen mit Hilfe der Google Directions API rekonstruiert. Mit den rekonstruierten Variablen Reisezeit und Reisekosten sowie verschiedenen sozioökonomischen und externen Variablen aus SrV 2008 wurden verschiedene Wahlmodelle geschätzt. Durch schrittweises Hinzufügen der Variablen konnte das Modell immer weiter verbessert werden. Wie zu erwarten, erwiesen sich dabei die Reisezeit und die Reisekosten als hoch signifikant. Von den restlichen Variablen waren jedoch lediglich das Geschlecht der befragten Person sowie die Wettersituation zum Zeitpunkt der Wahlentscheidung signifikant. Für das finale Modell wurden Zeitkostensätze errechnet und mit verschiedenen europäischen Studien verglichen. Die errechneten Zeitkostensätze erwiesen sich dabei als plausibel. Die SrV-Daten eignen sich also für die Schätzung von Wahlmodellen. / The following thesis analyzes, if revealed preference data from the German mobility survey "Mobilität in Städten" SrV 2008 is suited to estimate mode choice models. For that purpose, trip data from the survey was used and the different choice alternatives were reconstructed with the Google Directions API. Several mode choice models were estimated with the help of the reconstructed variables travel time and travel costs plus several socioeconomic and external variables from SrV 2008. The variables were added to the model step by step, thereby the quality of the model improved. As expected, travel time and travel costs were highly significant. However from the remaining variables only the gender of the person and the weather at the time of the trip were significant. For the final model, values of time were calculated and these were compared with values from different European studies. The calculated values of time proved to be feasible. Therefore, SrV data is suited to be used for mode choice models.

Die vorliegende Studienarbeit beschäftigt sich mit der Fragestellung, inwiefern sich Revealed-Preference-Daten aus der deutschen Mobilitätsbefragung "Mobilität in Städten" SrV 2008 dazu eignen, um basierend auf denen im Datensatz enthaltenen Wegen Verkehrsmittelwahlmodelle zu schätzen. Dazu wurden Wegedaten aus der Befragung verwendet, und die Wahlalternativen mit Hilfe der Google Directions API rekonstruiert. Mit den rekonstruierten Variablen Reisezeit und Reisekosten sowie verschiedenen sozioökonomischen und externen Variablen aus SrV 2008 wurden verschiedene Wahlmodelle geschätzt. Durch schrittweises Hinzufügen der Variablen konnte das Modell immer weiter verbessert werden. Wie zu erwarten, erwiesen sich dabei die Reisezeit und die Reisekosten als hoch signifikant. Von den restlichen Variablen waren jedoch lediglich das Geschlecht der befragten Person sowie die Wettersituation zum Zeitpunkt der Wahlentscheidung signifikant. Für das finale Modell wurden Zeitkostensätze errechnet und mit verschiedenen europäischen Studien verglichen. Die errechneten Zeitkostensätze erwiesen sich dabei als plausibel. Die SrV-Daten eignen sich also für die Schätzung von Wahlmodellen.:1 Einleitung 1 1.1 Vorstellung des Themas 1 1.2 Ziel dieser Arbeit 1 1.3 Gliederung dieser Arbeit 2 1.4 Wesentliche Ergebnisse 2 2 Theoretischer Hintergrund 3 2.1 Diskrete Wahltheorie 3 2.1.1 Deterministischer Nutzen 4 2.1.2 Stochastischer Störterm 5 2.1.3 Logit-Modell 6 2.2 Parameterschätzung 8 2.2.1 t-Test 9 2.2.2 Likelihood-Ratio Test 9 2.2.3 Likelihood-Ratio-Index 10 2.3 Datenquellen 10 3 Generierung eines RP-Datensatzes aus SrV-Daten 13 3.1 SrV 2008 13 3.2 Auswahl an Variablen und Datensätzen 13 3.3 Rekonstruktion der generischen Variablen 15 3.3.1 Google Directions API 15 3.3.2 Automatisierte API-Abfrage 16 3.3.3 Reisekosten 17 3.4 Erzeugter Datensatz 18 4 Modellentwicklung und Parameterschätzung 21 4.1 Entwicklung des Wahlmodells 21 4.2 Parameterschätzung mit Biogeme 24 4.3 Anwendung der Parameterschätzung auf die SrV-Daten 28 5 Diskussion der Modellergebnisse 33 5.1 Darstellung der Nutzeneinflüsse 33 5.2 Zeitkostensätze 35 5.3 Fehlerquellen 37 5.4 Fazit 38 Literaturverzeichnis 41 Datenquellen 45 Anhang 49 / The following thesis analyzes, if revealed preference data from the German mobility survey "Mobilität in Städten" SrV 2008 is suited to estimate mode choice models. For that purpose, trip data from the survey was used and the different choice alternatives were reconstructed with the Google Directions API. Several mode choice models were estimated with the help of the reconstructed variables travel time and travel costs plus several socioeconomic and external variables from SrV 2008. The variables were added to the model step by step, thereby the quality of the model improved. As expected, travel time and travel costs were highly significant. However from the remaining variables only the gender of the person and the weather at the time of the trip were significant. For the final model, values of time were calculated and these were compared with values from different European studies. The calculated values of time proved to be feasible. Therefore, SrV data is suited to be used for mode choice models.:1 Einleitung 1 1.1 Vorstellung des Themas 1 1.2 Ziel dieser Arbeit 1 1.3 Gliederung dieser Arbeit 2 1.4 Wesentliche Ergebnisse 2 2 Theoretischer Hintergrund 3 2.1 Diskrete Wahltheorie 3 2.1.1 Deterministischer Nutzen 4 2.1.2 Stochastischer Störterm 5 2.1.3 Logit-Modell 6 2.2 Parameterschätzung 8 2.2.1 t-Test 9 2.2.2 Likelihood-Ratio Test 9 2.2.3 Likelihood-Ratio-Index 10 2.3 Datenquellen 10 3 Generierung eines RP-Datensatzes aus SrV-Daten 13 3.1 SrV 2008 13 3.2 Auswahl an Variablen und Datensätzen 13 3.3 Rekonstruktion der generischen Variablen 15 3.3.1 Google Directions API 15 3.3.2 Automatisierte API-Abfrage 16 3.3.3 Reisekosten 17 3.4 Erzeugter Datensatz 18 4 Modellentwicklung und Parameterschätzung 21 4.1 Entwicklung des Wahlmodells 21 4.2 Parameterschätzung mit Biogeme 24 4.3 Anwendung der Parameterschätzung auf die SrV-Daten 28 5 Diskussion der Modellergebnisse 33 5.1 Darstellung der Nutzeneinflüsse 33 5.2 Zeitkostensätze 35 5.3 Fehlerquellen 37 5.4 Fazit 38 Literaturverzeichnis 41 Datenquellen 45 Anhang 49

Bus is the main mode of urban transport in most cities in developing countries. Despite a high mode share, bus service quality is often poor and para-transit services are regarded as a problem in urban transport systems rather than a solution. Using Dhaka as a case study, this thesis investigates bus service quality through identification and valuation of thirteen important attributes using discrete choice models. The attributes examined are travel time, travel cost, waiting time, headway, priority seats for women, crowding inside the bus, boarding and alighting, picking up and dropping off passengers, bus stop facilities, driving quality, driver and crew behaviour, cleanliness inside the bus, and air conditioning. Five focus groups were conducted to identify key qualitative bus attributes and their levels in order to design choice experiments for valuation. A survey of 431 respondents in Dhaka was then undertaken. Two choice experiments were designed and implemented within the survey, each with seven attributes (set A and set B) with travel cost as the common attribute. Multinomial Logit (MNL) models and Mixed Logit (MXL) models were developed using the Dhaka choice data. Twelve of the thirteen attributes were statistically significant at the 99% level. The values of in-vehicle time (IVT), waiting time and headway were BDT 34.80, 47.40 and 64.20 per hour respectively for low income groups in the segmented model. Waiting time has a premium valuation, 1.36 times higher than IVT, which endorses existing evidence. The highest valuation is for the dummy variable seating all the way which is BDT 42.20 for high income females. The next largest was bus stops properly, picks and drops passengers nicely , followed by wide door and mild steps for boarding and alighting , smooth and safe journey , bus stop with shed, but no seating arrangements , and air conditioning . The lowest value was BDT 4.61 for deck and seats are clean and tidy , for the low income group. The WTP for the qualitative attributes is high, but given the poor level of the existing service and low fare levels this seems reasonable. Income has a significant impact on travel cost, as well as gender on priority seats for women and crowding inside the bus. However, household car ownership does not have a significant impact on any of the bus attributes examined. The high income group has 75% higher WTP for A set attributes and 79% higher WTP for B set attributes than low income group. Females have 76% higher WTP for standing comfortably all the way , but 38% higher WTP for seating all the way compared to the male. However, females have a WTP of BDT 0.44 for per percent of priority seats for women in contrast with males who have a WTP of BDT -0.11. There is significant taste heterogeneity for both quantitative and qualitative attributes. The qualitative attributes for picking up and dropping off passengers, boarding and alighting facilities and driving facilities have higher valuation and this attributes came from the existing within the market competition structure in a highly fragmented bus market. Therefore, it is recommended to introduce competition for the market and incentives for bus industry consolidation.

A large body of literature studies the issues of the option price and other ex-ante welfare measures under the microeconomic theory to valuate reductions of risks inherent in environment and human health. However, it does not offer a careful discussion of how to estimate risk reduction values using data, especially the modeling and estimating individual perceptions of risks present in the econometric models. The central theme of my dissertation is the approaches taken for the empirical estimation of probabilistic risks under alternative assumptions about individual perceptions of risk involved: the objective probability, the Savage subjective probability, and the subjective distributions of probability. Each of these three types of risk specifications is covered in one of the three essays. The first essay addresses the problem of empirical estimation of individual willingness to pay for recreation access to public land under uncertainty. In this essay I developed an econometric model and applied it to the case of lottery-rationed hunting permits. The empirical result finds that the model correctly predicts the responses of 84% of the respondents in the Maine moose hunting survey. The second essay addresses the estimation of a logit model for individual binary choices that involve heterogeneity in subjective probabilities. For this problem, I introduce the use of the hierarchical Bayes to estimate, among others, the parameters of distribution of subjective probabilities. The Monte Carlo study finds the estimator asymptotically unbiased and efficient. The third essay addresses the problem of modeling perceived mortality risks from arsenic concentrations in drinking water. I estimated a formal model that allows for ambiguity about risk. The empirical findings revealed that perceived risk was positively associated with exposure levels and also related individuating factors, in particular smoking habits and one’s current health status. Further evidence was found that the variance of the perceived risk distribution is non-zero. In all, the three essays contribute methodological approaches and provide empirical examples for developing empirical models and estimating value of risk reductions in environment and human health, given the assumption about the individual’s perceptions of risk, and accordingly, the reasonable specifications of risks involved in the models.