scispace - formally typeset
Search or ask a question

Showing papers on "STAR model published in 2016"


Proceedings Article
01 Jan 2016
TL;DR: A new type of normalizing flow, inverse autoregressive flow (IAF), is proposed that, in contrast to earlier published flows, scales well to high-dimensional latent spaces and significantly improves upon diagonal Gaussian approximate posteriors.
Abstract: The framework of normalizing flows provides a general strategy for flexible variational inference of posteriors over latent variables. We propose a new type of normalizing flow, inverse autoregressive flow (IAF), that, in contrast to earlier published flows, scales well to high-dimensional latent spaces. The proposed flow consists of a chain of invertible transformations, where each transformation is based on an autoregressive neural network. In experiments, we show that IAF significantly improves upon diagonal Gaussian approximate posteriors. In addition, we demonstrate that a novel type of variational autoencoder, coupled with IAF, is competitive with neural autoregressive models in terms of attained log-likelihood on natural images, while allowing significantly faster synthesis.

901 citations


01 Jan 2016
TL;DR: Perhaps you have knowledge that, people have look hundreds of times for their chosen books like this likelihood based inference in cointegrated vector autoregressive models, but end up in harmful downloads.
Abstract: Thank you very much for downloading likelihood based inference in cointegrated vector autoregressive models. Maybe you have knowledge that, people have look hundreds times for their chosen books like this likelihood based inference in cointegrated vector autoregressive models, but end up in harmful downloads. Rather than reading a good book with a cup of coffee in the afternoon, instead they cope with some malicious bugs inside their desktop computer.

735 citations


Posted Content
TL;DR: This paper proposed a new type of normalizing flow, inverse autoregressive flow (IAF), that, in contrast to earlier published flows, scales well to high-dimensional latent spaces, and demonstrated that a novel type of variational autoencoder, coupled with IAF, is competitive with neural autoregression models in terms of attained log-likelihood on natural images, while allowing significantly faster synthesis.
Abstract: The framework of normalizing flows provides a general strategy for flexible variational inference of posteriors over latent variables. We propose a new type of normalizing flow, inverse autoregressive flow (IAF), that, in contrast to earlier published flows, scales well to high-dimensional latent spaces. The proposed flow consists of a chain of invertible transformations, where each transformation is based on an autoregressive neural network. In experiments, we show that IAF significantly improves upon diagonal Gaussian approximate posteriors. In addition, we demonstrate that a novel type of variational autoencoder, coupled with IAF, is competitive with neural autoregressive models in terms of attained log-likelihood on natural images, while allowing significantly faster synthesis.

193 citations


Journal ArticleDOI
TL;DR: In this paper, a novel test for unit root in general transitional autoregressive models, based on the infimum of t-ratios for the coefficient of a parametrized transition function, is proposed.
Abstract: This article develops a novel test for a unit root in general transitional autoregressive models, which is based on the infimum of t-ratios for the coefficient of a parametrized transition function. Our test allows for very flexible specifications of the transition function and short-run dynamics and is significantly more powerful than all the other existing tests. Moreover, we develop a large sample theory general enough to deal with randomly drifting parameter spaces, which is essential to properly test for a unit root against stationary transitional models. An empirical application of our test to the exchange rate data is also provided.

95 citations


Journal ArticleDOI
TL;DR: A simulation study is performed to compare the performance of three Inverse-Wishart prior specifications when one or more variances for the random effects in the multilevel autoregressive model are small, and shows that the prior specification that uses plug-in ML estimates of the variances performs best.
Abstract: Multilevel autoregressive models are especially suited for modeling between-person differences in within-person processes. Fitting these models with Bayesian techniques requires the specification of prior distributions for all parameters. Often it is desirable to specify prior distributions that have negligible effects on the resulting parameter estimates. However, the conjugate prior distribution for covariance matrices-the Inverse-Wishart distribution-tends to be informative when variances are close to zero. This is problematic for multilevel autoregressive models, because autoregressive parameters are usually small for each individual, so that the variance of these parameters will be small. We performed a simulation study to compare the performance of three Inverse-Wishart prior specifications suggested in the literature, when one or more variances for the random effects in the multilevel autoregressive model are small. Our results show that the prior specification that uses plug-in ML estimates of the variances performs best. We advise to always include a sensitivity analysis for the prior specification for covariance matrices of random parameters, especially in autoregressive models, and to include a data-based prior specification in this analysis. We illustrate such an analysis by means of an empirical application on repeated measures data on worrying and positive affect.

73 citations


Journal ArticleDOI
TL;DR: In this article, the performance of the monthly economic policy uncertainty (EPU) index in predicting recessionary regimes of the (quarterly) U.S. GDP was analyzed, and the authors applied a mixed-frequency Markov-switching vector autoregressive (MF-MSVAR) model.
Abstract: This paper analyzes the performance of the monthly economic policy uncertainty (EPU) index in predicting recessionary regimes of the (quarterly) U.S. GDP. In this regard, the authors apply a mixed-frequency Markov-switching vector autoregressive (MF-MSVAR) model, and compare its in-sample and out-of-sample forecasting performances to those of a Markov-switching vector autoregressive model (MS-VAR, where the EPU is averaged over the months to produce quarterly values) and a Markov-switching autoregressive (MS-AR) model. The results show that the MFMS-VAR fits the different recession regimes, and provides out-of-sample forecasts of recession probabilities which are more accurate than those derived from the MS-VAR and MS-AR models. The results highlight the importance of using high-frequency values of the EPU, and not averaging them to obtain quarterly values, when forecasting recessionary regimes for the U.S. economy.

63 citations


Journal ArticleDOI
TL;DR: Experimental results show that the evolutionary hybrid system presented promising results in the forecasting domain using a hybrid evolutionary system composed by a simple exponential smoothing filter, ARIMA and autoregressive (AR) linear models and a SVR model.

55 citations


Journal ArticleDOI
TL;DR: K-fold and Monte Carlo cross-validation and aggregation and aggregation (crogging) for combining neural network autoregressive forecasts demonstrate significant improvements in forecasting accuracy especially for short time series and long forecast horizons.

52 citations


Journal ArticleDOI
TL;DR: A multilevel threshold autoregressive model is proposed that can be used to detect state-dependent regulation with adequate power and Type I error and is illustrated with two empirical applications that extend the basic model to address additional substantive research questions.
Abstract: Intensive longitudinal data provide rich information, which is best captured when specialized models are used in the analysis. One of these models is the multilevel autoregressive model, which psychologists have applied successfully to study affect regulation as well as alcohol use. A limitation of this model is that the autoregressive parameter is treated as a fixed, trait-like property of a person. We argue that the autoregressive parameter may be state-dependent, for example, if the strength of affect regulation depends on the intensity of affect experienced. To allow such intra-individual variation, we propose a multilevel threshold autoregressive model. Using simulations, we show that this model can be used to detect state-dependent regulation with adequate power and Type I error. The potential of the new modeling approach is illustrated with two empirical applications that extend the basic model to address additional substantive research questions.

50 citations


Posted Content
TL;DR: This paper addresses the inference of the autoregressive parameters and associated network structure within a generalized linear model framework that includes Poisson and Bernoulli autore progressive processes and at the heart of this analysis is a sparsity-regularized maximum likelihood estimator.
Abstract: Vector autoregressive models characterize a variety of time series in which linear combinations of current and past observations can be used to accurately predict future observations. For instance, each element of an observation vector could correspond to a different node in a network, and the parameters of an autoregressive model would correspond to the impact of the network structure on the time series evolution. Often these models are used successfully in practice to learn the structure of social, epidemiological, financial, or biological neural networks. However, little is known about statistical guarantees on estimates of such models in non-Gaussian settings. This paper addresses the inference of the autoregressive parameters and associated network structure within a generalized linear model framework that includes Poisson and Bernoulli autoregressive processes. At the heart of this analysis is a sparsity-regularized maximum likelihood estimator. While sparsity-regularization is well-studied in the statistics and machine learning communities, those analysis methods cannot be applied to autoregressive generalized linear models because of the correlations and potential heteroscedasticity inherent in the observations. Sample complexity bounds are derived using a combination of martingale concentration inequalities and modern empirical process techniques for dependent random variables. These bounds, which are supported by several simulation studies, characterize the impact of various network parameters on estimator performance.

44 citations


Journal ArticleDOI
TL;DR: A class of generalized Poisson autoregressive models that properly capture flexible asymmetric and nonlinear responses through a switching mechanism are considered that can cope with data having a large portion of zeros and changes in dynamics.

Journal ArticleDOI
TL;DR: Applications to five MSCI stock market indices and to a large dataset of daily stock returns of all constituents of the Dax 30 illustrate the usefulness of the proposed model class in-sample and for density forecasting.
Abstract: We consider the problem of modeling the dependence among many time series. We build high-dimensional time-varying copula models by combining pair-copula constructions with stochastic autoregressive copula and generalized autoregressive score models to capture dependence that changes over time. We show how the estimation of this highly complex model can be broken down into the estimation of a sequence of bivariate models, which can be achieved by using the method of maximum likelihood. Further, by restricting the conditional dependence parameter on higher cascades of the pair copula construction to be constant, we can greatly reduce the number of parameters to be estimated without losing much flexibility. Applications to five MSCI stock market indices and to a large dataset of daily stock returns of all constituents of the Dax 30 illustrate the usefulness of the proposed model class in-sample and for density forecasting. Copyright © 2016 John Wiley & Sons, Ltd.

Proceedings Article
01 Jan 2016
TL;DR: In experiments with natural images, it is demonstrated that autoregressive flow leads to significant performance gains and is well applicable to models with high-dimensional latent spaces, such as convolutional generative models.
Abstract: We propose a simple and scalable method for improving the flexibility of variational inference through a transformation with autoregressive neural networks. Autoregressive neural networks, such as RNNs or the PixelCNN, are very powerful models and potentially interesting for use as variational posterior approximation. However, ancestral sampling in such networks is a long sequential operation, and therefore typically very slow on modern parallel hardware, such as GPUs. We show that by inverting autoregressive neural networks we can obtain equally powerful posterior models from which we can sample efficiently on modern hardware. We show that such data transformations, inverse autoregressive flows (IAF), can be used to transform a simple distribution over the latent variables into a much more flexible distribution, while still allowing us to compute the resulting variables' probability density function. The method is simple to implement, can be made arbitrarily flexible and, in contrast with previous work, is well applicable to models with high-dimensional latent spaces, such as convolutional generative models. The method is applied to a novel deep architecture of variational auto-encoders. In experiments with natural images, we demonstrate that autoregressive flow leads to significant performance gains.

Journal ArticleDOI
TL;DR: The Gaussian mixture vector autoregressive (GMVAR) model as mentioned in this paper is a mixture VAR model that is designed for analyzing time series that exhibit regime-switching behavior.

Journal ArticleDOI
TL;DR: In this paper, a score-based test for a double autoregressive model with conditional heteroscedastic (ARCH) errors was proposed, and two portmanteau-type statistics were derived for checking the adequacy of fitted model when either the error is nonnormal or the threshold is unknown.
Abstract: This article first proposes a score-based test for a double autoregressive model against a threshold double autoregressive (AR) model. It is an asymptotically distribution-free test and is easy to implement in practice. The article further studies the quasi-maximum likelihood estimation of a threshold double autoregressive model. It is shown that the estimated threshold is n-consistent and converges weakly to a functional of a two-sided compound Poisson process and the remaining parameters are asymptotically normal. Our results include the asymptotic theory of the estimator for threshold AR models with autoregressive conditional heteroscedastic (ARCH) errors and threshold ARCH models as special cases, each of which is also new in literature. Two portmanteau-type statistics are also derived for checking the adequacy of fitted model when either the error is nonnormal or the threshold is unknown. Simulation studies are conducted to assess the performance of the score-based test and the estimator in finite sa...


Journal ArticleDOI
TL;DR: In this paper, a model for analyzing time series of functions, subject to equality and inequality constraints at the two edges of the domain, respectively, such as daily demand and offer curves, is proposed.

Journal ArticleDOI
TL;DR: This paper proposes the Taylor Expansion Forecasting model as an alternative to the SVM and develops a novel hybrid methodology via combining autoregressive integrated moving average and Taylor expansion Forecasting to exploit the comprehensive forecasting capacity to the financial time series data with noise.
Abstract: Financial time series prediction is regarded as one of the most challenging job because of its inherent complexity, and the hybrid forecasting model incorporating autoregressive integrated moving average and support vector machine SVM has been implemented widely to deal with the both linear and nonlinear patterns in time series data. However, the SVM model does not take into consideration the time correlation knowledge between different data points in time series, which impacts the learning efficiency of the SVM in real application. To overcome this restriction, this paper proposes the Taylor Expansion Forecasting model as an alternative to the SVM and develops a novel hybrid methodology via combining autoregressive integrated moving average and Taylor Expansion Forecasting to exploit the comprehensive forecasting capacity to the financial time series data with noise. Both theoretical proof and empirical results obtained on several commodity future prices demonstrate that the proposed hybrid model improves greatly the forecasting accuracy.

Journal ArticleDOI
TL;DR: In this paper, the authors consider interpretation of estimates from the heterogeneous coefficient spatial autoregressive panel model of Aquaro et al. (2015) and derive partial derivatives (marginal effects) for this model.

Journal ArticleDOI
TL;DR: In this article, a convolutional functional autoregressive model is proposed, where the function at time t is a result of the sum of convolutions of the past functions and a set of convolution functions, plus a noise process.

Journal ArticleDOI
TL;DR: A simple stochastic process for modeling improper or noncircular complex-valued signals, extended to include a widely linear autoregressive term, which can then capture elliptical oscillations in a bivariate signal.
Abstract: We propose a simple stochastic process for modeling improper or noncircular complex-valued signals. The process is a natural extension of a complex-valued autoregressive process, extended to include a widely linear autoregressive term. This process can then capture elliptical, as opposed to circular, stochastic oscillations in a bivariate signal. The process is order one and is more parsimonious than alternative stochastic modeling approaches in the literature. We provide conditions for stationarity, and derive the form of the covariance and relation sequence of this model. We describe how parameter estimation can be efficiently performed both in the time and frequency domain. We demonstrate the practical utility of the process in capturing elliptical oscillations that are naturally present in seismic signals.

Journal ArticleDOI
TL;DR: In this paper, the authors proposed a new model for time series of counts, the autoregressive conditional negative binomial model that has a time-varying conditional auto-gressive mean function and heteroskedasticity.

Journal ArticleDOI
TL;DR: In this paper, the state-of-the-art ensemble model output statistics (EMOS) is combined with an ensemble that is adjusted by an autoregressive process fitted to the respective error series by a spread-adjusted linear pool in the case of temperature forecasts.
Abstract: To address the uncertainty in outputs of numerical weather prediction (NWP) models, ensembles of forecasts are used. To obtain such an ensemble of forecasts, the NWP model is run multiple times, each time with variations in the mathematical representations of the model and/or initial or boundary conditions. To correct for possible biases and dispersion errors in the ensemble, statistical postprocessing models are frequently employed. These statistical models yield full predictive probability distributions for a weather quantity of interest and thus allow for a more accurate representation of forecast uncertainty. This article proposes to combine the state-of-the-art Ensemble Model Output Statistics (EMOS) with an ensemble that is adjusted by an autoregressive process fitted to the respective error series by a spread-adjusted linear pool in the case of temperature forecasts. The basic ensemble modification technique we introduce may be used to simply adjust the ensemble itself as well as to obtain a full predictive distribution for the weather quantity. As demonstrated for temperature forecasts from the European Centre for Medium-Range Weather Forecasts ensemble, the proposed procedure gives rise to improved results over the basic (local) EMOS method.

Journal ArticleDOI
TL;DR: It is demonstrated that the non-Gaussian time series with a generalized Laplace marginal distribution has the ability to accurately represent hilliness features of road topography providing a significant improvement over a purely Gaussian model.

Journal ArticleDOI
TL;DR: In this article, the authors address the issue of parameter dimensionality reduction in vector autoregressive models (VARs) for many variables by imposing specific reduced rank restrictions on the coefficient matrices that simplify the VARs into Multivariate Autoregressive Index (MAI).

Journal ArticleDOI
TL;DR: Two alternative approaches that can be implemented using Gibbs sampling methods in a straightforward way and which allow one to deal with the problem of model uncertainty in spatial autoregressive models in a flexible and computationally efficient way are presented.
Abstract: This paper compares the performance of Bayesian variable selection approaches for spatial autoregressive models. It presents two alternative approaches that can be implemented using Gibbs sampling methods in a straightforward way and which allow one to deal with the problem of model uncertainty in spatial autoregressive models in a flexible and computationally efficient way. A simulation study shows that the variable selection approaches tend to outperform existing Bayesian model averaging techniques in terms of both in-sample predictive performance and computational efficiency. The alternative approaches are compared in an empirical application using data on economic growth for European NUTS-2 regions.

Journal ArticleDOI
TL;DR: In this article, a group of logistic smooth transition heterogeneous autoregressive (LSTHAR) models of realized volatility is proposed, which can simultaneously approximate long memory behavior and describe sign and size asymmetries.

Dissertation
09 Feb 2016
TL;DR: In this paper, the authors present models that can efficiently analyse long univariate time series and large (possibly) unbalanced panels of time series, and a variety of applications are presented to show the performance of the newly developed models.
Abstract: 2 This dissertation studies time varying parameter models for discrete valued time series. The models under consideration are the class of state space models and the generalized autoregressive score models. It presents models that can efficiently analyse long univariate time series and large (possibly) unbalanced panels of time series. A variety of applications are presented to show the performance of the newly developed models. The number of goals that are scored and conceded by football teams in the English Premier League and the German Bundesliga are analysed. Another prominent example in this dissertation is the extraction of volatility from discrete stock price changes and the intraday dependence structures between the series.

Journal ArticleDOI
TL;DR: In this paper, the forecasting performance of various ARIMA models by using time series data was compared by using the Box-Jenkins approach for forecasting, and two important models for forecasting out of many existing were classified.
Abstract: This study compares the forecasting performance of various Autoregressive integrated moving average (ARIMA) models by using time series data. Primarily, The Box-Jenkins approach is considered here for forecasting. For empirical analysis, we used CPI as a proxy for inflation and employed quarterly data from 1970 to 2006 for Pakistan. The study classified two important models for forecasting out of many existing by taking into account various initial steps such as identification, the order of integration and test for comparison. However, later model 2 turn out to be a better model than model 1 after considering forecasted errors and the number of comparative statistics.

Journal ArticleDOI
TL;DR: In this paper, quantile regression for a wide class of time series models including Comparisons among weakly (ARMA) models with asymmetric generalized autoregressive conditional heteroscedasticity errors is considered.
Abstract: This paper considers quantile regression for a wide class of time series models including Comparisons among weakly (ARMA) models with asymmetric generalized autoregressive conditional heteroscedasticity errors. The classical mean-variance models are reinterpreted as conditional location-scale models so that the quantile regression method can be naturally geared into the considered models. The consistency and asymptotic normality of the quantile regression estimator is established in location-scale time series models under mild conditions. In the application of this result to ARMA-generalized autoregressive conditional heteroscedasticity models, more primitive conditions are deduced to obtain the asymptotic properties. For illustration, a simulation study and a real data analysis are provided.