The discovery of generalized causal models with mixed variables using MML criterion

One major difficulty frustrating the application of linear causal models is that they are not easily adapted to cope with discrete data. This is unfortunate since most real problems involve both continuous and discrete variables. In this paper, we consider a class of graphical models which allow both continuous and discrete variables, and propose the parameter estimation method and a structure discovery algorithm based on Minimum Message Length and parameter estimation. Experimental results are given to demonstrate the potential for the application of this method.

Internationalization and alliance formation : evidence from Turkish SMEs

This study explores the issue of internationalization through forming alliances with foreign capital in the small business sector in Turkey. Using a sample of 257 SMEs from this emerging market economy, collected via a field study, it finds that Turkish SMEs would like to form alliances with foreign capital for expanding their production capacity and accessing to world markets. Multivariate multinomial logit models are employed to analyse the survey data econometrically. Size-specific, sector-specific and management-specific factors are identified in the alliance motivation. The `market' and `finance' aspects of alliances prevail in the multivariate analysis with significant implications. There is also evidence that some conditional relationships offered by multivariate analysis differ from unconditional associations found in the survey, which implies that alliance motivation is partly a product of multidimensional decision-making process on behalf of SMEs.

Is there a unit root in the inflation rate? New evidence from panel data models with multiple structural breaks

In this article, we examine whether or not the inflation rate for 17 OECD countries can be modelled as a stationary process. We find that (1) conventional univariate unit root tests without any structural breaks generally reveal that the inflation rate contains a unit root; (2) the KPSS univariate test with multiple structural breaks reveals that for 10 out of 17 countries inflation is stationary; and (3) the KPSS panel unit root test reveals strong evidence for stationarity of the inflation rate for panels consisting of countries which were declared nonstationary by univariate tests.

Dynamic interactions in models with Kaleckian lags

Provides a careful assessment of previous research on lags in economic models. Several interesting lines of research are opened up. Chief among them is the analysis of bubbles and their bursting in the financial components of economic models.

Hierarchical hidden Markov models with general state hierarchy

The hierarchical hidden Markov model (HHMM) is an extension of the hidden Markov model to include a hierarchy of the hidden states. This form of hierarchical modeling has been found useful in applications such as handwritten character recognition, behavior recognition, video indexing, and text retrieval. Nevertheless, the state hierarchy in the original HHMM is restricted to a tree structure. This prohibits two different states from having the same child, and thus does not allow for sharing of common substructures in the model. In this paper, we present a general HHMM in which the state hierarchy can be a lattice allowing arbitrary sharing of substructures. Furthermore, we provide a method for numerical scaling to avoid underflow, an important issue in dealing with long observation sequences. We demonstrate the working of our method in a simulated environment where a hierarchical behavioral model is automatically learned and later used for recognition.

Building monotonicity-preserving fuzzy inference models with optimization-based similarity reasoning and a monotonicity index

In this paper, a novel approach to building a Fuzzy Inference System (FIS) that preserves the monotonicity property is proposed. A new fuzzy re-labeling technique to re-label the consequents of fuzzy rules in the database (before the Similarity Reasoning process) and a monotonicity index for use in FIS modeling are introduced. The proposed approach is able to overcome several restrictions in our previous work that uses mathematical conditions in building monotonicity-preserving FIS models. Here, we show that the proposed approach is applicable to different FIS models, which include the zero-order Sugeno FIS and Mamdani models. Besides, the proposed approach can be extended to undertake problems related to the local monotonicity property of FIS models. A number of examples to demonstrate the usefulness of the proposed approach are presented. The results indicate the usefulness of the proposed approach in constructing monotonicity-preserving FIS models.

Testing for unit roots in panel time-series models with multiple level breaks

This paper proposes two new unit root tests that are appropriate in the presence of an unknown number of structural breaks in the level of the data. One is based on a single time series and the other is based on a panel of multiple series. For the estimation of the number of breaks and their locations, a simple procedure based on outlier detection is proposed. The limiting distributions of the tests are derived and evaluated in small samples using simulation experiments. The implementation of the tests is illustrated using as an example purchasing power parity.

A factor analytical method to interactive effects dynamic panel models with or without unit root

In a recent study, Bai (Fixed-Effects Dynamic Panel Models, A Factor Analytical Method. Econometrica 81, 285-314, 2013a) proposes a new factor analytic (FA) method to the estimation of dynamic panel data models, which has the unique and very useful property that it is completely bias-free. However, while certainly appealing, it is restricted to fixed effects models without a unit root. In many situations of practical relevance this is a rather restrictive consideration. The purpose of the current study is therefore to extend the FA approach to cover models with multiple interactive effects and a possible unit root.

Small-variance asymptotics for bayesian nonparametric models with constraints

The users often have additional knowledge when Bayesian nonparametric models (BNP) are employed, e.g. for clustering there may be prior knowledge that some of the data instances should be in the same cluster (must-link constraint) or in different clusters (cannot-link constraint), and similarly for topic modeling some words should be grouped together or separately because of an underlying semantic. This can be achieved by imposing appropriate sampling probabilities based on such constraints. However, the traditional inference technique of BNP models via Gibbs sampling is time consuming and is not scalable for large data. Variational approximations are faster but many times they do not offer good solutions. Addressing this we present a small-variance asymptotic analysis of the MAP estimates of BNP models with constraints. We derive the objective function for Dirichlet process mixture model with constraints and devise a simple and efficient K-means type algorithm. We further extend the small-variance analysis to hierarchical BNP models with constraints and devise a similar simple objective function. Experiments on synthetic and real data sets demonstrate the efficiency and effectiveness of our algorithms.

Robust adaptive backstepping excitation controller design for simple power system models with external disturbances

This paper presents a nonlinear robust adaptive excitation controller design for a simple power system model where a synchronous generator is connected to an infinite bus. The proposed controller is designed to obtain the adaption laws for estimating critical parameters of synchronous generators which are considered as unknown while providing the robustness against the bounded external disturbances. The convergence of different physical quantities of a single machine infinite bus (SMIB) system, with the proposed control scheme, is ensured through the negative definiteness of the derivative of Lyapunov functions. The effects of external disturbances are considered during formulation of Lyapunov function and thus, the proposed excitation controller can ensure the stability of the SMIB system under the variation of critical parameters as well as external disturbances including noises. Finally, the performance of the proposed scheme is investigated with the inclusion of external disturbances in the SMIB system and its superiority is demonstrated through the comparison with an existing robust adaptive excitation controller. Simulation results show that the proposed scheme provides faster responses of physical quantities than the existing controller.

Returns to scale, market power, and the nature of price rigidity in new keynesian models with self-fulfilling expectations

Increasing returns to scale and firms' market power are two potential sources of sunspot expectations in neoclassical models. We show that in New Keynesian models, returns to scale and market power can have fundamentally different implications for broad macroeconomic issues, including self-fulfilling expectations, depending on the nature of price rigidity. Our findings suggest that the design of stabilization monetary policy can depend on precise knowledge about the economy's real and nominal features. Therefore, a clear understanding of the specific economic environment and its relevance to monetary policymaking for ensuring macroeconomic stability can be an integrated part of monetary policy practice.

Dirichlet process mixture models with pairwise constraints for data clustering

The Dirichlet process mixture (DPM) model, a typical Bayesian nonparametric model, can infer the number of clusters automatically, and thus performing priority in data clustering. This paper investigates the influence of pairwise constraints in the DPM model. The pairwise constraint, known as two types: must-link (ML) and cannot-link (CL) constraints, indicates the relationship between two data points. We have proposed two relevant models which incorporate pairwise constraints: the constrained DPM (C-DPM) and the constrained DPM with selected constraints (SC-DPM). In C-DPM, the concept of chunklet is introduced. ML constraints are compiled into chunklets and CL constraints exist between chunklets. We derive the Gibbs sampling of the C-DPM based on chunklets. We further propose a principled approach to select the most useful constraints, which will be incorporated into the SC-DPM. We evaluate the proposed models based on three real datasets: 20 Newsgroups dataset, NUS-WIDE image dataset and Facebook comments datasets we collected by ourselves. Our SC-DPM performs priority in data clustering. In addition, our SC-DPM can be potentially used for short-text clustering.

Changes in the rates of weight and waist circumference gain in Australian adults over time: a longitudinal cohort study

OBJECTIVE: To assess in a single cohort whether annual weight and waist circumference (WC) change has varied over time.

DESIGN: Longitudinal cohort study with three surveys (1) 1999/2000; (2) 2004/2005 and (3) 2011/2012. Generalised linear mixed models with random effects were used to compare annualised weight and WC change between surveys 1 and 2 (period 1) with that between surveys 2 and 3 (period 2). Models were adjusted for age to analyse changes with time rather than age. Models were additionally adjusted for sex, education status, area-level socioeconomic disadvantage, ethnicity, body mass index, diabetes status and smoking status.

SETTING: The Australian Diabetes, Obesity and Lifestyle study (AusDiab)-a population-based, stratified-cluster survey of 11247 adults aged ≥25 years. PARTICIPANTS: 3351 Australian adults who attended each of three surveys and had complete measures of weight, WC and covariates.

PRIMARY OUTCOME MEASURES: Weight and WC were measured at each survey. Change in weight and WC was annualised for comparison between the two periods.

RESULTS: Mean weight and WC increased in both periods (0.34 kg/year, 0.43 cm/year period 1; 0.13 kg/year, 0.46 cm/year period 2). Annualised weight gain in period 2 was 0.11 kg/year (95% CI 0.06 to 0.15) less than period 1. Lesser annual weight gain between the two periods was not seen for those with greatest area-level socioeconomic disadvantage, or in men over the age of 55. In contrast, the annualised WC increase in period 2 was greater than period 1 (0.07 cm/year, 95% CI 0.01 to 0.12). The increase was greatest in men aged 55+ years and those with a greater area-level socioeconomic disadvantage.

CONCLUSIONS: Between 2004/2005 and 2011/2012, Australian adults in a national study continued to gain weight, but more slowly than 1999/2000-2004/2005. While weight gain may be slowing, this was not observed for older men or those in more disadvantaged groups, and the same cannot be said for WC.