The effects of random guessing as a measurement disturbance on Rasch fit statistics (unweighted total, weighted total, and unweighted ability between) and the Logit Residual Index (LRI) were examined through simulated data sets of varying sample sizes, test lengths, and distribution types. Three test lengths (25, 50, and 100), three sample sizes (25, 50, and 100), two item difficulty distributions (normal and uniform), and three levels of guessing (no guessing (0%), 25%, and 50%) were used in the simulations, resulting in 54 experimental conditions. The mean logit person ability for each experiment was +1. Each experimental condition was simulated once in an effort to approximate what could happen on the single administration of a four option per item multiple choice test to a group of relatively high ability persons. Previous research has shown that varying item and person parameters have no effect on Rasch fit statistics. Consequently, these parameters were used in the present study to establish realistic test conditions, but were not interpreted as effect factors in determining the results of this study.

"Cross-Border Contagion in the Banking Sector: The Case of Nordic Countries" by Lina Baronaite Abstract: The objective of the thesis is to estimate the degree of cross-border contagion among the Nordic banking sectors. It analyzes a sample of sixteen largest listed Nordic banks from January 2004 to January 2014. Using a multinomial logit model we test whether there is any degree of contagion among the four banking sectors, whether it is more pro- nounced for larger banks and whether the recent financial crisis has exacerbated it. Our results are in line with similar studies conducted for other countries. In particular, we find that a shock in one bank- ing sector is positively associated with an increase in shocks in another banking sector. Second, these shocks are larger and more significant for larger and more active international banks. Finally, the effect of the recent financial crisis has ambiguous effects on the cross-sectoral banking contagion. It appears that contagious links between some sec- tors weakened (Sweden and Denmark, Sweden and Finland). Other economies (Sweden and Norway) on the contrary became more depen- dent on each other. The results are robust to a wide variety of changes in specifications.

Charles University in Prague Faculty of Social Sciences Institute of Economic Studies MASTER THESIS The determinants of access to finance. Evidence for transition economies Author: Alina Cazachevici Supervisor: Roman Horvath, Ph.D. Academic Year: 2012/2013 Abstract The thesis provides an empirical analysis of impact of country-level and firm-level determinants on access to finance in transition economies. Generalized Ordered Logit model is applied on survey data for transition countries, combined with financial market indicators. The results show that higher concentration in banking sector, as well as higher financial deepening have a positive impact on access to finance, while volatile macroeconomic environment, higher implication of foreign-owned and state- owned banks seems to be perceived as increasing obstacles in accessing external financing. Combining indexes for liberalization in banking sector and liberalization of securities markets proved that before liberalization process firms had better access to finance. One of the possible explanations is that before liberalization state banks were forced by politicians to issue more loans, while after reforms the political pressure was removed, imposing stricter conditions for loan granting. Inclusion of corruption variable yields expectable results that...

Ziel dieser Arbeit ist die Untersuchung der Bedeutung der Spezifikation für Ratingmodelle zur Prognose von Kreditausfallwahrscheinlichkeiten. Ausgehend von dem in der Bankenpraxis etablierten Logit-Modell werden verschiedene Modellerweiterungen diskutiert und hinsichtlich ihrer Eigenschaften als Ratingmodelle empirisch und simulationsbasiert untersucht. Die Interpretierbarkeit und die Prognosegüte der Modelle werden dabei gleichermaßen berücksichtigt. Besonderes Augenmerk wird auf Mixed Logit-Modelle zur Abbildung individueller Heterogenität gelegt. Die Ergebnisse zeigen, dass die Spezifikation einen wichtigen Einfluss auf die Eigenschaften von Ratingmodellen hat und dass insbesondere mit Hilfe von Mixed Logit-Ansätzen sinnvoll interpretierbare Ratingmodelle mit guten Prognoseeigenschaften erlangt werden können. / This thesis aims at analyzing the impact of model specification on credit rating models as used by banks to forecast credit default probabilities. The well-established logit model is extended in various directions. Implications are discussed empirically and with the help of Monte Carlo studies. In this respect, interpretability and forecasting quality are considered simultaneously. A main focus is on mixed logit models that allow for modeling individual heterogeneity. Results indicate that proper model specification has a strong impact on the quality of rating models. In particular it is emphasized that mixed logit models offer well interpretable credit rating models with high forecasting power.

Demand for domestic tobacco has decreased over the past two decades. In 2004, the tobacco buyout program terminated marketing quotas and price support established under the federal tobacco program in 1938. Additionally, in 2003, the World Health Organization Framework Convention on Tobacco Control (WTO FCTC) acknowledged the importance of supply control in conjunction with demand control strategies to eliminate or reduce the consumption of tobacco products. According to the Census of Agriculture, the number of tobacco farms in the US fell by 40 percent between 2002 and 2007. Tobacco farmers are looking for alternative on-farm and off-farm sources of income. This study uses a rank-ordered logit model (ROLM) to explore factors affecting farmers’ perceptions about the potential for grain crops, cotton, peanuts, hay, fruits and vegetables, cow/calf, dairy, beef cattle and, other crops/livestock to replace tobacco production. Results suggest that hay is one of the on-farm enterprises perceived as having the highest potential to replace tobacco among burley tobacco farmers. Age, education, farm size and farm cash receipts were found to affect farmers’ perceptions about the potential for different alternative enterprises to replace tobacco. Additionally, results suggest that researchers should be careful when designing ranking questions in order to maximize rate of response and quality of the data obtained from this type of questions.

This work is concerned with the study of nonlinear nonconvex stochastic programming, in particular in the context of trust-region approaches. We first explore how to exploit the structure of multistage stochastic nonlinear programs with linear constraints, in the framework of primal-dual interior point methods. We next study consistency of sample average approximations (SAA) for general nonlinear stochastic programs. We also develop a new algorithm to solve the SAA problem, using the statistical inference information to reduce numercial costs, by means of an internal variable sample size strategy. We finally assess the numerical efficiency of the proposed method for the estimation of discrete choice models, more precisely mixed logit models, using our software AMLET, written for this purpose.

An economic literature concerns instruments to improve the preference elicitation methods for the reform-based governmental programs. We construct an instrument for the budget allocation method using a Cobb-Douglas functional form. We apply the instrument to the survey data which is collected for Swedish Recreational Fishing Industry to elicit the preferences for governmental management actions. We analyze the elasticity or weights in the instrument by the binary logit and censored regression models and by comparing the significant estimates by the gross and net effects we get results which increase credence to the instrument we apply. / I am heartily thankful to my supervisor, Thomas Laitila, whose guidance and support from the initial to the final level enabled me to develop the thesis.More, I offer my regards to Anders Lunander who supported me in any respect during the completion of the thesis.

The purpose of this paper is to investigate how individuals with different characteristics make their choice-decisions when consuming STIGA table tennis blades, which are combinations of various attributes, such as price, control, attack, etc. It is expected that the general trend of choice behavior on this special commodity can be, at least to some extent, revealed. Data were collected using questionnaires sent to registered members of a table tennis club in China. The questionnaires included information and questions about individuals’ monthly income levels, ages, technique styles, etc. A multinomial logit model was then applied to analyze factors determining Chinese consumers’ choice behavior on STIGA table tennis blades. The results indicated that the main element influencing Chinese consumers’ choice of STIGA ping-pong blades was the technique style and other variables did not seem to influence the choice of table tennis blades. These results might be explained by the limited sample size as well as unmeasured and immeasurable factors. Thus, a more extensive research is needed to be conducted in the future.

We follow the approach of expressive voting and consider that voters with more extreme ideology can enjoy higher utility after voting. However, along with effect of ideology, voters also take the difference of his ideology and candidate¡¦s into account. Given the above assumptions, two candidates choose their ideology before voters decide whether and for whom to vote. Two candidates¡¦ ideology converges to the middle point if voters¡¦ utility generated by expressive voting approach is less important than by instrumental voting approach. In the opposite, two candidates¡¦ ideology diverge and the voters with mild ideology refuse to vote in point if voters¡¦ utility generated by expressive voting approach is more important. We examine the ANES data of ten U.S. presidential elections from 1972 to 2008. The nested logit model is used to estimate the corresponding coefficients of voters¡¦ utility generated by expressive voting approach and instrumental voting approach. The data supports our conjecture that voters with more extreme ideology are more likely to cast their votes.

This dissertation contains two essays on the effect of welfare reform on child- birth, marriage, and divorce. In the first essay, I exploit the cross state variation in welfare reform implementation to identify its effect on birth rates. The results from multinomial logit models suggest that the welfare reform significantly increased the probability of marital births. The out-of-wedlock birth rates decreased but this effect is not significant. The strong work incentives decrease birth rates in both marital and non-marital statuses suggesting that bearing a child is not appealing for women who are more progressive in careers. However, the most aggressive welfare policy significantly increases marital birth as expected. Birth rates among teenage girls are not affected by the welfare reform. I further investigate the effect of the family cap policy. Using a semi-natural experiment, I compare the birth rate of women who already have had a second or higher order birth (treatment group) to women who have had one child (comparison group), in states with and without family caps. The difference in difference estimates reveal a strictly negative effect of family caps on the higher order birth rates as expected. In the second essay, I use reduced-form estimation and cross-state variation in timing of reform adoption to extract both mechanical and behavioral effects of welfare reform on marriage and divorce likelihood. I construct a flow measure of marriage and divorce by matching individuals in the Current Population Survey from March 1988B to 2002 and observing changes in marital status. I introduce a converse matching procedure to detect women who are not in the survey for two consecutive years. I find that the welfare reform has a significantly negative effect on marriage rates and an insignificant effect on divorce rates. The Difference-in-Difference estimates suggest that marriage among disadvantaged women is negatively affected by the welfare reform. I also provide a theoretical model to decompose the effect of welfare reform on marriage due to each of the components of the reform, i.e., time limits, work sanctions, earnings disregards, and maximum cash benefits. My results provide a novel explanation for the effects of work incentives and welfare restrictions on marriage.