|
on Econometrics |
| By: | T. Arduini |
| Abstract: | This paper proposes a semiparametric estimator for spatial autoregressive (SAR) binary choice models in the context of panel data with fixed effects. The estimation procedure is based on the observational equivalence between distribution free models with a conditional median restriction and parametric models (such as Logit/Probit) exhibiting (multiplicative) heteroskedasticity and autocorrelation. Without imposing any parametric structure on the error terms, we consider the semiparametric nonlinear least squares (NLLS) estimator for this model and analyze its asymptotic properties under spatial near-epoch dependence. The main advantage of our method over the existing estimators is that it consistently estimates choice probabilities. The finite-dimensional estimator is shown to be consistent and root-n asymptotically normal under some reasonable conditions. Finally, a Monte Carlo study indicates that the estimator performs quite well in finite samples. |
| JEL: | C14 C21 C23 C25 R15 |
| Date: | 2016–01 |
| URL: | http://d.repec.org/n?u=RePEc:bol:bodewp:wp1052&r=ecm |
| By: | Zacharias Psaradakis (University of London); Marian Vavra (National Bank of Slovakia, Research Department) |
| Abstract: | This paper considers the problem of testing for linearity of stationary time series. Portmanteau tests are discussed which are based on generalized correlations of residuals from a linear model (that is, autocorrelations and cross-correlations of different powers of the residuals). The finite-sample properties of the tests are assessed by means of Monte Carlo experiments. The tests are applied to 100 time series of stock returns. |
| Keywords: | autocorrelation; cross-correlation; nonlinearity; Portmanteau test; stock returns |
| JEL: | C12 C22 C52 |
| Date: | 2016–05 |
| URL: | http://d.repec.org/n?u=RePEc:svk:wpaper:1037&r=ecm |
| By: | Smith Paul (Department of Economics, University of Strathclyde) |
| Abstract: | Nowcasting UK GDP during the Depression reviews the performance of several statistical techniques in nowcasting preliminary estimates of UK GDP, particularly during the recent depression. Traditional bridging equations, MIDAS regressions and factor models are all considered. While there are various theoretical differences and perceived advantages for each technique, replicated real-time out-o-ample testing shows that, in practice, there is in fact little to choose between methods in terms of end-of-period nowcasting accuracy. The analysis also reveals that none of the aforementioned statistical models can consistently beat a consensus of professional economists in nowcasting preliminary GDP estimates. This inability of statistical models to beat the consensus may reflect several factors, one of which is the revisions and re-appraisal of rends inherent in UK GDP statistics. The suggestion is that these changes impact on observed relationships between GDP and indicator variables such as business surveys, which impairs nowcasting performance. Indeed, using a synthetic series based purely on observed preliminary GDP estimates, which introduces stability to the target variable series, the nowcasting accuracy of regressions including closely-watched PMI data is improved by 25-40 percentage points relative to a naive benchmark. |
| Keywords: | Nowcasting, Forecasting, Real-time data, GDP, MIDAS |
| JEL: | C22 C32 |
| Date: | 2016–04 |
| URL: | http://d.repec.org/n?u=RePEc:str:wpaper:1606&r=ecm |
| By: | Adam Clements (QUT); Ayesha Scott (QUT); Annastiina Silvennoinen (QUT) |
| Abstract: | This paper explores the link between equicorrelation and market volatility. The standard equicorrelation model is extended to condition the correlation process on volatility, based on the Volatility Dependent Dynamic Conditional Correlation class of model. Analysis of this relationship is presented in two empirical examples, with both a national and international context studied. The various correlation forecasting methods are compared using a portfolio allocation problem, specifically the global minimum variance portfolio and Model Confidence Set. Relative economic value is also considered. In the case of U.S. equities, overall the equicorrelation models perform well and the inclusion of volatility in the equicorrelations performs well against the standard equicorrelated model. For large portfolios a simple specification such as constant conditional correlation seems sufficient, particularly during periods of market calm. Internationally, the equicorrelated models perform poorly against the dynamic conditional corelation-based models. Reasoning is provided that the information pooling advantage equicorrelation has over dynamic conditional correlation models is eroded when forecasting correlations between indices, rather than equities. In both applications, there appears to be no statistically significant difference between the standard equicorrelation model and the Volatility Dependent class although in general a volatility dependent structure leads to lower portfolio variances. |
| Keywords: | Volatility, multivariate GARCH, equicorrelation, portfolio allocation |
| JEL: | C22 G11 G17 |
| Date: | 2016–05–11 |
| URL: | http://d.repec.org/n?u=RePEc:qut:auncer:2016_02&r=ecm |
| By: | Gouel, Christophe; Legrand, Nicolas |
| Keywords: | Commodity price dynamics, storage, Bayesian inference, Demand and Price Analysis, Research Methods/ Statistical Methods, C51, C52, Q11, |
| Date: | 2016 |
| URL: | http://d.repec.org/n?u=RePEc:ags:aaea16:235599&r=ecm |
| By: | Breitmoser, Yves |
| Abstract: | Multinomial logit is the canonical model of discrete choice and widely used to analyze preferences and welfare. Its axiomatic foundation is incomplete: binomial logit is assumed; independence of irrelevant alternatives extends logit to multinomial choice. The present paper provides a self-contained foundation, showing that the axiom "binomial choice is logit" is behaviorally founded by "narrow bracketing" and "no systematic mistakes" (e.g. default effects). This in turn allows me to drop the no-mistakes axiom, yielding generalized logit accounting for systematic mistakes axiomatically consistently. The results position logit in the "mistakes-debate" in behavioral welfare and clarify the foundation for the functional form. |
| Keywords: | logit, axiomatic foundation, discrete choice, utility estimation, welfare |
| JEL: | D03 |
| Date: | 2016–05–28 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:71632&r=ecm |