Testing the Prebish-Singer hypothesis using second-generation panel data stationarity tests with a break

In this paper, we test the Prebish-Singer (PS) hypothesis, which states that real commodity prices decline in the long run, using two recent powerful panel data stationarity tests accounting for cross-sectional dependence and a structural break. We find that the hypothesis cannot be rejected for most commodities other than oil.

Statistical Tests for Joint Analysis of Performance Measures

Recently there has been an increasing interest in the development of new methods using Pareto optimality to deal with multi-objective criteria (for example, accuracy and architectural complexity). Once one has learned a model based on their devised method, the problem is then how to compare it with the state of art. In machine learning, algorithms are typically evaluated by comparing their performance on different data sets by means of statistical tests. Unfortunately, the standard tests used for this purpose are not able to jointly consider performance measures. The aim of this paper is to resolve this issue by developing statistical procedures that are able to account for multiple competing measures at the same time. In particular, we develop two tests: a frequentist procedure based on the generalized likelihood-ratio test and a Bayesian procedure based on a multinomial-Dirichlet conjugate model. We further extend them by discovering conditional independences among measures to reduce the number of parameter of such models, as usually the number of studied cases is very reduced in such comparisons. Real data from a comparison among general purpose classifiers is used to show a practical application of our tests.

The Trading of Unlimited Liability Bank Shares in Nineteenth Century Ireland: The Bagehot Hypothesis

In the mid-1820s, banks became the first businesses in Great Britain and Ireland to be allowed to form freely on an unlimited liability joint-stock basis. Walter Bagehot warned that their shares would ultimately be owned by widows, orphans, and other impecunious individuals. Another hypothesis is that the governing bodies of these banks, constrained by special legal restrictions on share trading, acted effectively to prevent such shares being transferred to the less wealthy. We test both conjectures using the archives of an Irish joint-stock bank. The results do not support Bagehot's hypothesis.

Panel unit root tests in the presence of cross-sectional dependence: finite sample performance and an application

This paper examines the finite sample properties of three testing regimes for the null hypothesis of a panel unit root against stationary alternatives in the presence of cross-sectional correlation. The regimes of Bai and Ng (2004), Moon and Perron (2004) and Pesaran (2007) are assessed in the presence of multiple factors and also other non-standard situations. The behaviour of some information criteria used to determine the number of factors in a panel is examined and new information criteria with improved properties in small-N panels proposed. An application to the efficient markets hypothesis is also provided. The null hypothesis of a panel random walk is not rejected by any of the tests, supporting the efficient markets hypothesis in the financial services sector of the Australian Stock Exchange.

Association Study of 167 Candidate Genes for Schizophrenia Selected by a Multi-Domain Evidence-Based Prioritization Algorithm and Neurodevelopmental Hypothesis

Integrating evidence from multiple domains is useful in prioritizing disease candidate genes for subsequent testing. We ranked all known human genes (n = 3819) under linkage peaks in the Irish Study of High-Density Schizophrenia Families using three different evidence domains: 1) a meta-analysis of microarray gene expression results using the Stanley Brain collection, 2) a schizophrenia protein-protein interaction network, and 3) a systematic literature search. Each gene was assigned a domain-specific p-value and ranked after evaluating the evidence within each domain. For comparison to this
ranking process, a large-scale candidate gene hypothesis was also tested by including genes with Gene Ontology terms related to neurodevelopment. Subsequently, genotypes of 3725 SNPs in 167 genes from a custom Illumina iSelect array were used to evaluate the top ranked vs. hypothesis selected genes. Seventy-three genes were both highly ranked and involved in neurodevelopment (category 1) while 42 and 52 genes were exclusive to neurodevelopment (category 2) or highly ranked (category 3), respectively. The most significant associations were observed in genes PRKG1, PRKCE, and CNTN4 but no individual SNPs were significant after correction for multiple testing. Comparison of the approaches showed an excess of significant tests using the hypothesis-driven neurodevelopment category. Random selection of similar sized genes from two independent genome-wide association studies (GWAS) of schizophrenia showed the excess was unlikely by chance. In a further meta-analysis of three GWAS datasets, four candidate SNPs reached nominal significance. Although gene ranking using integrated sources of prior information did not enrich for significant results in the current experiment, gene selection using an a priori hypothesis (neurodevelopment) was superior to random selection. As such, further development of gene ranking strategies using more carefully selected sources of information is warranted.

Non-enzymatic glycosylation of skin collagen in patients with type 1 (insulin-dependent) diabetes mellitus and limited joint mobility

Forearm skin biopsies were obtained from diabetic subjects with and without limited joint mobility, and from non-diabetic control subjects. Collagen purified from these samples was assayed for non-enzymatic glycosylation. The level in all diabetic patients was significantly greater than that in control subjects (p less than 0.001), but those diabetic patients with limited joint mobility had a level of collagen glycosylation similar to that in those with normal joints (15.3 +/- 1.3 and 16.5 +/- 1.3 nmol fructose/10 mg protein, respectively; mean +/- SEM). Glycosylation of collagen in the diabetic patients correlated with glycosylated haemoglobin measured at the time of skin biopsy (r = 0.60). These results do not support the hypothesis that non-enzymatic glycosylation of collagen, as reflected by the ketoamine link, plays an important role in the development of limited joint mobility in diabetes.

Accuracy of shoulder joint and subacromial injections in cadavers.

OBJECTIVE:

"Blind" shoulder injections are often inaccurate and infiltrate untargeted structures. We tested a hypothesis that optimizing certain anatomical and positional factors would improve accuracy and reduce dispersal.

METHODS:

We evaluated one subacromial and one glenohumeral injection technique on cadavers.

RESULTS:

Mean accuracy was 91% for subacromial-targeted and 74 and 91% (worst- and best-case scenarios) for joint-targeted injections. Mean dispersal was 19% for subacromial-targeted and 16% for joint-targeted injections. All results bettered those reported previously.

CONCLUSION:

These "optimized" techniques might improve accuracy and limit dispersal of blind shoulder injections in clinical situations, benefiting efficacy and safety. However, evaluation is required in a clinical setting.

Testing the Prebisch–Singer hypothesis since 1650: Evidence from panel techniques that allow for multiple breaks

In this paper, we re-examine two important aspects of the dynamics of relative primary commodity prices, namely the secular trend and the short run volatility. To do so, we employ 25 series, some of them starting as far back as 1650 and powerful panel data stationarity tests that allow for endogenous multiple structural breaks. Results show that all the series are stationary after allowing for endogenous multiple breaks. Test results on the Prebisch–Singer hypothesis, which states that relative commodity prices follow a downward secular trend, are mixed but with a majority of series showing negative trends. We also make a first attempt at identifying the potential drivers of the structural breaks. We end by investigating the dynamics of the volatility of the 25 relative primary commodity prices also allowing for endogenous multiple breaks. We describe the often time-varying volatility in commodity prices and show that it has increased in recent years.

Synergy Between an Improved Covariate Unit Root Test and Cross-sectionally Dependent Panel Data Unit Root Tests, with two Applications

This paper proposes the use of an improved covariate unit root test which exploits the cross-sectional dependence information when the panel data null hypothesis of a unit root is rejected. More explicitly, to increase the power of the test, we suggest the utilization of more than one covariate and offer several ways to select the ‘best’ covariates from the set of potential covariates represented by the individuals in the panel. Employing our methods, we investigate the Prebish-Singer hypothesis for nine commodity prices. Our results show that this hypothesis holds for all but the price of petroleum.

Bayesian Dependence Tests for Continuous, Binary and Mixed Continuous-Binary Variables

Tests for dependence of continuous, discrete and mixed continuous-discrete variables are ubiquitous in science. The goal of this paper is to derive Bayesian alternatives to frequentist null hypothesis significance tests for dependence. In particular, we will present three Bayesian tests for dependence of binary, continuous and mixed variables. These tests are nonparametric and based on the Dirichlet Process, which allows us to use the same prior model for all of them. Therefore, the tests are “consistent” among each other, in the sense that the probabilities that variables are dependent computed with these tests are commensurable across the different types of variables being tested. By means of simulations with artificial data, we show the effectiveness of the new tests.

Joint Analysis of Multiple Algorithms and Performance Measures

There has been an increasing interest in the development of new methods using Pareto optimality to deal with multi-objective criteria (for example, accuracy and time complexity). Once one has developed an approach to a problem of interest, the problem is then how to compare it with the state of art. In machine learning, algorithms are typically evaluated by comparing their performance on different data sets by means of statistical tests. Standard tests used for this purpose are able to consider jointly neither performance measures nor multiple competitors at once. The aim of this paper is to resolve these issues by developing statistical procedures that are able to account for multiple competing measures at the same time and to compare multiple algorithms altogether. In particular, we develop two tests: a frequentist procedure based on the generalized likelihood-ratio test and a Bayesian procedure based on a multinomial-Dirichlet conjugate model. We further extend them by discovering conditional independences among measures to reduce the number of parameters of such models, as usually the number of studied cases is very reduced in such comparisons. Data from a comparison among general purpose classifiers is used to show a practical application of our tests.