Search

Proxy means testing (PMT) and community-based targeting (CBT) are two of the leading methods for targeting social assistance in developing countries. In this paper, we present a hybrid targeting method that incorporates CBT's emphasis on local information and preferences with PMT's reliance on verifiable indicators. Specifically, we outline a Bayesian framework for targeting that resembles PMT in that beneficiary selection is based on a weighted sum of sociodemographic characteristics. We nevertheless propose calibrating the weights to preference rankings from community targeting exercises, implying that the weights used by our method reflect how potential beneficiaries themselves substitute sociodemographic features when making targeting decisions. We discuss several practical extensions to the model, including a generalization to multiple rankings per community, an adjustment for elite capture, a method for incorporating auxiliary information on potential beneficiaries, and a dynamic updating procedure. We further provide an empirical illustration using data from Burkina Faso and Indonesia.

In the context of a vector autoregression (VAR) model, or any multivariate regression model, the number of relevant predictors may be small relative to the information set available from which to build a prediction equation. It is well known that forecasts based off of (un-penalized) least squares estimates can overfit the data and lead to poor predictions. Since the Minnesota prior was proposed (Doan et al. (1984)), there have been many methods developed aiming at improving prediction performance. In this paper we propose the horseshoe prior (Carvalho et al. (2010), Carvalho et al. (2009)) in the context of a Bayesian VAR. The horseshoe prior is a unique shrinkage prior scheme in that it shrinks irrelevant signals rigorously to 0 while allowing large signals to remain large and practically unshrunk. In an empirical study, we show that the horseshoe prior competes favorably with shrinkage schemes commonly used in Bayesian VAR models as well as with a prior that imposes true sparsity in the coefficient vector. Additionally, we propose the use of particle Gibbs with backwards simulation (Lindsten et al. (2012), Andrieu et al. (2010)) for the estimation of the time-varying volatility parameters. We provide a detailed description of all MCMC methods used in the supplementary material that is available online.

In statistical modeling we strive to specify models that resemble data collected in studies or observed from processes. Consequently, distributional specification and parameter estimation are central to parametric models. Graphical procedures, such as the quantile-quantile (Q-Q) plot, are arguably the most widely used method of distributional assessment, though critics find their interpretation to be overly subjective. Formal goodness-of-fit tests are available and are quite powerful, but only indicate whether there is a lack of fit, not why there is lack of fit. In this paper we explore the use of the lineup protocol to inject rigor to graphical distributional assessment and compare its power to that of formal distributional tests. We find that lineups of standard Q-Q plots are more powerful than lineups of de-trended Q-Q plots and that lineup tests are more powerful than traditional tests of normality. While, we focus on diagnosing non-normality, our approach is general and can be directly extended to the assessment of other distributions., Comment: 21 pages, 9 figures, 3 tables