In a discrete choice experiment, each respondent typically chooses the best product or service sequentially from many groups or choice sets of alternatives which are characterized by a number of different attributes. Respondents can find it difficult to trade off prospective products or services when every attribute of the offering changes in each comparison. Especially in studies involving many attributes, respondents get overloaded by the complexity of the choice task. To overcome respondent fatigue, it makes sense to simplify the comparison by holding some of the attributes constant in every choice set. A study in the health care
literature where eleven attributes were allocated across three different experimental designs with only five attributes being varied motivates the approach we present. However, our algorithmic approach is more general,allowing for any number of attributes and a smaller number of fixed attributes.

In this paper, we present a general discrete-time affine framework aimed at jointly modeling yield curves associated with different debtors. The underlying fixed-income securities may differ in terms of credit quality and/or in terms of liquidity. The risk factors follow conditionally Gaussian processes, with drifts and variance-covariance matrices that are subject to regime shifts described by a Markov chain with (historical) non-homogenous transition probabilities. While flexible, the model remains tractable. In particular, bond prices are given by quasi-explicit formulas. Various numerical examples are proposed, including a sector-contagion model and credit-rating modeling.

Every econometrician/statistician is familiar with quantiles and percentiles. In combination with penalized splines they can be used to fit smooth curves to the outer regions of a scatterplots. The core of conditional quantile estimation is an objective function based on a weighted sum of absolute values of residuals, with weights that depend on their signs. If we give positive residuals weight 0.9 and negative residuals 0.1, we aim at the 90-th percentile curve. We get very interesting results if we replace the absolute values by squares, but keep the asymmetric weighting. Then we are estimating conditional expectile curves. Expectiles are unknown to the majority of statisticians, but they are useful and fascinating. They also are easy to compute. I will introduce expectiles, their use in smoothing, and their relationship to the (conditional) data density. A common problem with quantile and (to a lesser degree) expectile smoothing is that estimated curves for different asymmetries cross each other. Theoretically this cannot happen, but in practice it does, and it is a nuisance. I will discuss several solutions.

The Brethren of the Common Life (BCL) were a religious community founded by Geert Groote in the city of Deventer in the Netherlands in the late fourteenth century. The BCL stimulated the accumulation of human capital through schools, libraries and the production of books and laid the intellectual foundation for Christian Humanism in the Low Countries and the rest of Northern Europe. This paper empirically investigates the long-lasting impact of the BCL on a range of economic and societal outcomes in the Low Countries. Our empirical estimates provide evidence that the BCL has contributed to the uniquely and early high rates of literacy in the Low Countries in the decades before the amazing development of the Dutch Republic in the seventeenth century. In addition, we find positive effects of the BCL on early book production and on city growth in the fifteenth and sixteenth century. Finally, we find that cities with BCL-roots significantly earlier joined the Dutch Revolt against the Habsburg Rulers. These findings are supported by regressions that use distance to Deventer as an instrument for the presence of BCL.

Estimation of the Mixed Multinomial Logit model using permuted antithetic draws

Abstract
Estimation of the mixed multinomial logit (MMNL) model requires the use of maximum simulated likelihood (MSL) to locate the parameters of interest. The simulation of the random parameter distributions is typically done using quasi Monte Carlo (QMC) techniques. In this paper, we demonstrate that the econometric structure of the MMNL model will result in multiple MSL solutions given a finite number of QMC draws when the most predominate type of simulated draws currently employed within the literature are used. We then demonstrate how the use of full factorial antithetic draws can be used to provide a single unique solution.

and

Can we separate scale and coefficient heterogeneity in random coefficients models?

Abstract
There is growing interest in the notion that a significant component of the heterogeneity retrieved in random coefficients models may actually relate to variations in absolute sensitivities, i.e. scale heterogeneity, rather than variations in relative sensitivities. This has in part motivated the development of specialised modelling tools such as the G-MNL model. While not disagreeing with the notion that scale heterogeneity across respondents exists, this paper argues that attempts in the literature to disentangle scale from preference heterogeneity are not possible given that the two effects are perfectly confounded. In particular, we show how the various model specifications can in fact simply be seen as different parameterisations, and that any gains in fit are the results of using more flexible distributions, rather than an ability to capture scale heterogeneity. We illustrate our arguments through an empirical example and show how the conclusions from past work are based on misinterpretations of model results.

Robust Sparse Principal Components Analysis

Abstract
A method for principal component analysis is proposed that is sparse and robust at the same time. The sparsity delivers principal components that have loadings on a small number of variables, making them easier to interpret. The robustness makes the analysis resistant to outlying observations. The principal components correspond to directions that maximize a robust measure of the variance, with an additional penalty term to take sparseness into account. We propose an algorithm to compute the sparse and robust principal components. The method is applied on several real data examples, and diagnostic plots for detecting outliers and for selecting the degree of sparsity are provided. A simulation experiment studies the effect on statistical efficiency by requiring both robustness and sparsity.

and

Discriminating between fractional noise and spurious long memory

Abstract
Many time series in economics and finance seem to exhibit long memory in the sense that the autocorrelations decline rather slowly. It is commonplace to model such persistence by fractionally integrated processes. However, it has been pointed out that many other types of processes that are indeed non-fractional can produce long memory. Examples includes time series processes with breaks and structural change as well as many non-linear time series models. It is of interest to distinguish between these different types of long memory processes since correct specification is essential for inference, model-building and forecasting. In this paper we exploit the fact that non-linear transformations of fractional noise processes will themselves exhibit long memory of a lower degree than the original series. In fact, the limiting degree of long memory is dictated by the Hermite rank of the corresponding non-linear transformation. The feature can be used as a device for discriminating fractional processes from other classes of long memory models.