|
on Econometrics |
| By: | Mark E. McGovern (Harvard School of Public Health); Till Bärnighausen (Harvard School of Public Health); Joshua A. Salomon (Harvard School of Public Health); David Canning (Harvard School of Public Health) |
| Abstract: | Selection bias in HIV prevalence estimates occurs if refusal to test is correlated with HIV status. Interviewer identity is plausibly correlated with consenting to test, but not with HIV status, allowing a Heckman-type correction that produces consistent HIV prevalence estimates. We innovate on existing approaches by showing that an interviewer random effects Bayesian estimator produces prevalence estimates that are unbiased as well as consistent. An additional advantage of this new estimator is that it allows the construction of bootstrapped standard errors. It is also easily implemented in standard statistical software. The model is used to produce new estimates and confidence intervals for HIV prevalence among men in Zambia and Ghana. |
| Keywords: | HIV, Heckman Selection Models, Missing Data, Bayesian Estimation |
| Date: | 2013–04 |
| URL: | http://d.repec.org/n?u=RePEc:gdm:wpaper:10113&r=ecm |
| By: | Helmut Lütkepohl (DIW, FU Berlin); Anna Staszewska-Bystrova (University of Lodz); Peter Winker (University of Lodz) |
| Abstract: | In vector autoregressive analysis confidence intervals for individual impulse responses are typically reported to indicate the sampling uncertainty in the estimation results. A range of methods are reviewed and a new proposal is made for constructing joint confidence bands, given a pre-specified coverage level, for the impulse responses at all horizons considered simultaneously. The methods are compared in a simulation experiment and recommendations for empirical work are provided. |
| Keywords: | Vector autoregressive process, impulse responses, bootstrap, confidence band |
| JEL: | C32 |
| Date: | 2013 |
| URL: | http://d.repec.org/n?u=RePEc:mar:magkse:201325&r=ecm |
| By: | Ito, Ryoko |
| Abstract: | A spline-DCS model is developed to forecast the conditional distribution of high-frequency financial data with periodic behavior. The dynamic cubic spline of Harvey and Koopman (1993) is applied to allow diurnal patterns to evolve stochastically over time. An empirical application illustrates the practicality and impressive predictive performance of the model. |
| Keywords: | outlier; robustness, score, calendar effect, spline, trade volume. |
| JEL: | C22 |
| Date: | 2013–04–19 |
| URL: | http://d.repec.org/n?u=RePEc:cam:camdae:1315&r=ecm |
| By: | Marc Hallin; Chintan Mehta |
| Keywords: | independent component analysis (ICA); local asymptotic normality (LAN); ranks; R-Estimation; Robustness |
| Date: | 2013–04 |
| URL: | http://d.repec.org/n?u=RePEc:eca:wpaper:2013/142876&r=ecm |
| By: | Kociecki, Andrzej |
| Abstract: | The paper aims at systematic placement of identification concept within Bayesian approach. Pointing to some deficiencies of the standard Bayesian language to describe identification problem we propose several useful characterizations that seem to be intuitively sound and attractive given their potential applications. We offer comprehensive interpretations for them. Moreover we introduce the concepts of uniform, marginal and faithful identification. We argue that all these concepts may have practical significance. Our theoretical development is illustrated with a number of simple examples and one real application i.e. Structural VAR model. |
| Keywords: | Bayesian, Identification |
| JEL: | C01 C11 C51 |
| Date: | 2013–04–25 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:46538&r=ecm |
| By: | Lukasz Gatarek (Econometric Institute, Erasmus School of Economics, Erasmus University Rotterdam); Lennart Hoogerheide (VU University Amsterdam); Koen Hooning (Delft University of Technology); Herman K. van Dijk (Econometric Institute, Erasmus University Rotterdam, and VU University Amsterdam) |
| Abstract: | Accurate prediction of risk measures such as Value at Risk (VaR) and Expected Shortfall (ES) requires precise estimation of the tail of the predictive distribution. Two novel concepts are introduced that offer a specific focus on this part of the predictive density: the censored posterior, a posterior in which the likelihood is replaced by the censored likelihood; and the censored predictive likelihood, which is used for Bayesian Model Averaging. We perform extensive experiments involving simulated and empirical data. Our results show the ability of these new approaches to outperform the standard posterior and traditional Bayesian Model Averaging techniques in applications of Value-at-Risk prediction in GARCH models. |
| Keywords: | censored likelihood; censored posterior; censored predictive likelihood; Bayesian Model Averaging; Value at Risk; Metropolis-Hastings algorithm. |
| JEL: | C11 C15 C22 C51 C53 C58 G17 |
| Date: | 2013–04–15 |
| URL: | http://d.repec.org/n?u=RePEc:dgr:uvatin:20130060&r=ecm |
| By: | Paola Cerchiello (Department of Economics and Management, University of Pavia); Paolo Giudici (Department of Economics and Management, University of Pavia) |
| Abstract: | The measurement of the quality of academic research is a rather controversial issue. Recently Hirsch has proposed a measure that has the advantage of summarizing in a single summary statistics all the information that is contained in the citation counts of each scientist. From that seminal paper, a huge amount of research has been lavished, focusing on one hand on the development of correction factors to the h index and on the other hand, on the pros and cons of such measure proposing several possible alternatives. Although the h index has received a great deal of interest since its very beginning, only few papers have analyzed its statistical properties and implications, typically from an asymptotic viewpoint. In the present work we propose an exact statistical approach to derive the distribution of the h index. To achieve this objective we work directly on the two basic components of the h index: the number of produced papers and the related citation counts vector, by introducing convolution models. Our proposal is applied to a database of homogeneous scientists made up of 131 full professors of statistics employed in Italian universities. The results show that while ”sufficient” authors are reasonably well detected by a crude bibliometric approach, outstanding ones are underestimated, motivating the development of a statistical based h index. Our proposal offers such development and in particular exact confidence intervals to compare authors as well as quality control thresholds that can be used as target values. |
| Date: | 2013–04 |
| URL: | http://d.repec.org/n?u=RePEc:pav:demwpp:039&r=ecm |
| By: | Peter Fuleky (UHERO and Department of Economics, University of Hawaii at Manoa); Carl S. Bonham (Department of Economics, University of Hawaii at Manoa) |
| Abstract: | We analyze the forecasting performance of small mixed frequency factor models when the observed variables share stochastic trends. The indicators are observed at various frequencies and are tied together by cointegration so that valuable high fre- quency information is passed to low frequency series through the common factors. Dierencing the data breaks the cointegrating link among the series and some of the signal leaks out to the idiosyncratic components, which do not contribute to the trans- fer of information among indicators. We nd that allowing for common trends improves forecasting performance over a stationary factor model based on dierenced data. The \common-trends factor model" outperforms the stationary factor model at all analyzed forecast horizons. Our results demonstrate that when mixed frequency variables are cointegrated, modeling common stochastic trends improves forecasts. |
| Keywords: | Dynamic Factor Model, Mixed Frequency Samples, Common Trends, Forecasting |
| JEL: | E37 C32 C53 L83 |
| Date: | 2013–04 |
| URL: | http://d.repec.org/n?u=RePEc:hai:wpaper:201305&r=ecm |
| By: | Duran-Vazquez, Rocio; Lorenzo-Valdes, Arturo; Ruiz-Porras, Antonio |
| Abstract: | We develop a GARCH model with autoregressive conditional asymmetry to describe time-series. This means that, in addition to the conditional mean and variance, we assume that the skewness describes the behavior of the time-series. Analytically, we use the methodology proposed by Fernández and Steel (1998) to define the behavior of the innovations of the model. We use the approach developed by Brooks, et. al., (2005), to build it. Moreover, we show its usefulness by modeling the daily returns of the Mexican Stock Market Index (IPC) during the period between January 3rd, 2008 and September 29th, 2009. |
| Keywords: | Conditional Asymmetry; GARCH; Skewness; Stock Market Returns; Mexico |
| JEL: | C22 G10 |
| Date: | 2013–04–17 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:46328&r=ecm |
| By: | Kociecki, Andrzej |
| Abstract: | We provide some generalization and clarification of the identification conditions for Structural VAR (SVAR) models given in Rubio–Ramírez et al (2010). In particular we show that their basic sufficient condition is also necessary. In addition we give necessary and sufficient conditions for identification almost everywhere in SVAR under homogenous restrictions irrespective of whether the model is exactly identified or over–identified. The modification of the order condition is also suggested. |
| Keywords: | SVAR, identification |
| JEL: | C10 C32 E52 |
| Date: | 2013–04–25 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:46536&r=ecm |
| By: | Pierpaolo Pierani; Silvia Tiezzi |
| Abstract: | This paper deals with one of the main theoretical and empirical problems associated with the rational addiction model, namely that the demand equation derived from the rational addiction theory is not empirically distinguishable from models with forward-looking behavior, but with timeinconsistent preferences. The implication is that, even when forward-looking behavior can be convincingly supported, this equation cannot provide evidence in favor of time-consistent preferences against a model with dynamic inconsistency. In fact, we show that the possibility of testing for exponential versus non-exponential time discounting is nested within the rational addiction model. We propose a test of time consistency that uses only the information obtained from the general rational addiction demand equation and the price effects. A pseudo panel of Italian households is used to test for rational addiction in tobacco consumption. GMM estimators are used to deal with errors in variables and unobserved heterogeneity. The results conform to the theoretical predictions. We find evidence that tobacco consumers are forward-looking, but timeinconsistent. The values of the derived present bias and long run discount parameters are statistically significant and in line with the literature. |
| Keywords: | rational addiction, time inconsistency, GMM |
| JEL: | C23 D03 D12 |
| Date: | 2012–12 |
| URL: | http://d.repec.org/n?u=RePEc:usi:wpaper:666&r=ecm |
| By: | Marco M. Sorge |
| Abstract: | This note points out a hitherto unrecognised identification issue in a class of rational expectations (RE) models with news shocks. We show that different degrees of anticipation (information flows) have strikingly different implications for the identifiability of the underlying structural model, irrespective of its non-fundamental time-series representation. In particular, under full shock anticipation equilibrium reduced forms behave as noisy perfect foresight state motions, which are non-identifiable. As a consequence, the underlying news shocks model fails to be (first-order) identified. The identification failure is illustrated with a New Keynesian model that can be solved analytically. |
| Keywords: | Rational expectations, perfect foresight, news shocks, identification. |
| JEL: | C1 E32 |
| Date: | 2013–04–08 |
| URL: | http://d.repec.org/n?u=RePEc:eei:rpaper:eeri_rp_2013_08&r=ecm |
| By: | Kajornrit, Jesada; Wong, Kok Wai; Fung, Chun Che |
| Abstract: | Estimation of missing precipitation records is one of the most important tasks in hydro-logical and environmental study. The efficiency of hydrological and environmental models is sub-ject to the completeness of precipitation data. This study compared some basic soft computing techniques, namely, artificial neural network, fuzzy inference system and adaptive neuro-fuzzy in-ference system as well as the conventional methods to estimate missing monthly rainfall records in the northeast region of Thailand. Four cases studies are selected to evaluate the accuracy of the es-timation models. The simultaneous rainfall data from three nearest neighbouring control stations are used to estimate missing records at the target station. The experimental results suggested that the adaptive neuro-fuzzy inference system could be considered as a recommended technique because it provided the promising estimation results, the estimation mechanism is transparent to the users, and do not need prior knowledge to create the model. The results also showed that fuzzy inference system could provide compatible accuracy to artificial neural network.In addition, artificial neural network must be used with care becausesuch model is sensitive to irregular rainfall events. -- |
| Keywords: | Missing Precipitation Records,Artificial Neural Network,Fuzzy Inference System,Adaptive Neuro-Fuzzy Inference System,Northeast Region of Thailand |
| Date: | 2012 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:itsb12:72486&r=ecm |
| By: | M. Fatih Ekinci; Gazi Kabas; Enes Sunel |
| Abstract: | Estimating a robust and stable trend is an important challenge for economic analysis. We compare alternative approaches by estimating the cyclical component for the real exchange rate series of Turkey. Comparison criteria is the sensitivity of the estimated cycle to additional data points. A formal test reveals that cycle values obtained with all methods change substantially upon new data arrivals. To rank the performance of the methods, additional measures underlining the comovement of real-time cycles and the cyclical values with additional data, and the magnitude of end-point bias are developed. These criteria show that an unobserved components approach, which assumes trend and cycle innovations are orthogonal, and xes the share of trend shocks on the real depreciation rate fluctuations at 10 percent, dominates alternative filtering methods. |
| Keywords: | Trend-cycle decompositions, real exchange rates, stochastic trend |
| JEL: | C22 E37 F31 |
| Date: | 2013 |
| URL: | http://d.repec.org/n?u=RePEc:tcb:wpaper:1316&r=ecm |