Subsampling inference in threshold autoregressive models

This paper discusses inference in self exciting threshold autoregressive (SETAR)models. Of main interest is inference for the threshold parameter. It iswell-known that the asymptotics of the corresponding estimator depend uponwhether the SETAR model is continuous or not. In the continuous case, thelimiting distribution is normal and standard inference is possible. Inthe discontinuous case, the limiting distribution is non-normal and cannotbe estimated consistently. We show valid inference can be drawn by theuse of the subsampling method. Moreover, the method can even be extendedto situations where the (dis)continuity of the model is unknown. In thiscase, also the inference for the regression parameters of the modelbecomes difficult and subsampling can be used advantageously there aswell. In addition, we consider an hypothesis test for the continuity ofthe SETAR model. A simulation study examines small sample performance.

Estimating learning models from experimental data

We study the statistical properties of three estimation methods for a model of learning that is often fitted to experimental data: quadratic deviation measures without unobserved heterogeneity, and maximum likelihood withand without unobserved heterogeneity. After discussing identification issues, we show that the estimators are consistent and provide their asymptotic distribution. Using Monte Carlo simulations, we show that ignoring unobserved heterogeneity can lead to seriously biased estimations in samples which have the typical length of actual experiments. Better small sample properties areobtained if unobserved heterogeneity is introduced. That is, rather than estimating the parameters for each individual, the individual parameters are considered random variables, and the distribution of those random variables is estimated.

Convergence in panel data: Evidence from the skipping estimation

This paper demonstrates that, unlike what the conventional wisdom says, measurement error biases in panel data estimation of convergence using OLS with fixed effects are huge, not trivial. It does so by way of the "skipping estimation"': taking data from every m years of the sample (where m is an integer greater than or equal to 2), as opposed to every single year. It is shown that the estimated speed of convergence from the OLS with fixed effects is biased upwards by as much as 7 to 15%.

The economic effects of the Protestant Reformation: Testing the Weber hypothesis in the German Lands

Many theories, most famously Max Weber s essay on the Protestant ethic, have hypothesizedthat Protestantism should have favored economic development. With their considerablereligious heterogeneity and stability of denominational affiliations until the 19th century, theGerman Lands of the Holy Roman Empire present an ideal testing ground for this hypothesis.Using population figures in a dataset comprising 272 cities in the years 1300 1900, I find no effectsof Protestantism on economic growth. The finding is robust to the inclusion of a varietyof controls, and does not appear to depend on data selection or small sample size. In addition,Protestantism has no effect when interacted with other likely determinants of economic development.I also analyze the endogeneity of religious choice; instrumental variables estimates ofthe effects of Protestantism are similar to the OLS results.

Fish oil for prevention of sudden death in hemodialysis patients?

Friedman et al. report that hemodialysis patients with the highest levels of n-3 fatty acids had impressively low odds of sudden cardiac death. The study is limited by a small sample size, and the analysis relies on only a single baseline measurement of blood levels. Recent randomized evidence indeed fails to support that n-3 fatty acids may prevent sudden death in nonrenal patients. More evidence is needed to advocate fish oil in this setting.

Influence of chronic hyperglycemia on cerebral microvascular remodeling: an in vivo study using perfusion computed tomography in acute ischemic stroke patients.

BACKGROUND AND PURPOSE: To investigate the effect of chronic hyperglycemia on cerebral microvascular remodeling using perfusion computed tomography. METHODS: We retrospectively identified 26 patients from our registry of 2453 patients who underwent a perfusion computed tomographic study and had their hemoglobin A1c (HbA1c) measured. These 26 patients were divided into 2 groups: those with HbA1c>6.5% (n=15) and those with HbA1c≤6.5% (n=11). Perfusion computed tomographic studies were processed using a delay-corrected, deconvolution-based software. Perfusion computed tomographic values were compared between the 2 patient groups, including mean transit time, which relates to the cerebral capillary architecture and length. RESULTS: Mean transit time values in the nonischemic cerebral hemisphere were significantly longer in the patients with HbA1c>6.5% (P=0.033), especially in the white matter (P=0.005). Significant correlation (R=0.469; P=0.016) between mean transit time and HbA1c level was observed. CONCLUSIONS: Our results from a small sample suggest that chronic hyperglycemia may be associated with cerebral microvascular remodeling in humans. Additional prospective studies with larger sample size are required to confirm this observation.

A comparison of methods for differential expression analysis of RNA-seq data.

BACKGROUND: Finding genes that are differentially expressed between conditions is an integral part of understanding the molecular basis of phenotypic variation. In the past decades, DNA microarrays have been used extensively to quantify the abundance of mRNA corresponding to different genes, and more recently high-throughput sequencing of cDNA (RNA-seq) has emerged as a powerful competitor. As the cost of sequencing decreases, it is conceivable that the use of RNA-seq for differential expression analysis will increase rapidly. To exploit the possibilities and address the challenges posed by this relatively new type of data, a number of software packages have been developed especially for differential expression analysis of RNA-seq data. RESULTS: We conducted an extensive comparison of eleven methods for differential expression analysis of RNA-seq data. All methods are freely available within the R framework and take as input a matrix of counts, i.e. the number of reads mapping to each genomic feature of interest in each of a number of samples. We evaluate the methods based on both simulated data and real RNA-seq data. CONCLUSIONS: Very small sample sizes, which are still common in RNA-seq experiments, impose problems for all evaluated methods and any results obtained under such conditions should be interpreted with caution. For larger sample sizes, the methods combining a variance-stabilizing transformation with the 'limma' method for differential expression analysis perform well under many different conditions, as does the nonparametric SAMseq method.

Anticipatory pleasure skills training: a new intervention to reduce anhedonia in schizophrenia.

PURPOSE: Anhedonia is a challenging symptom of schizophrenia and remains largely recalcitrant to current pharmacological treatments. The goal of this exploratory pilot study was to assess if a cognitive-sensory intervention could improve anticipatory pleasure. DESIGN AND METHODS: Five participants meeting the Diagnostic and Statistical Manual of Mental Disorders (4th edition, Text Revision) criteria for schizophrenia, presenting severe anhedonia and stabilized on atypical antipsychotic medication, received between 10 hours and 25 hours of training. FINDINGS: Results show that the patients improved on the anticipatory scale of the Temporal Experience of Pleasure Scale. Daily activities of the patients were also increased. PRACTICE IMPLICATIONS: These preliminary data need to be interpreted with caution given the small sample of the study, but they offer promising paths to develop new interventions to alleviate anhedonia in schizophrenia.

The weaker points of fish acute toxicity tests and how tests on embryos can solve some issues.

Fish acute toxicity tests play an important role in environmental risk assessment and hazard classification because they allow for first estimates of the relative toxicity of various chemicals in various species. However, such tests need to be carefully interpreted. Here we shortly summarize the main issues which are linked to the genetics and the condition of the test animals, the standardized test situations, the uncertainty about whether a given test species can be seen as representative to a given fish fauna, the often missing knowledge about possible interaction effects, especially with micropathogens, and statistical problems like small sample sizes and, in some cases, pseudoreplication. We suggest that multi-factorial embryo tests on ecologically relevant species solve many of these issues, and we shortly explain how such tests could be done to avoid the weaker points of fish acute toxicity tests.

Level shifts in a panel data based unit root test : an application to the rate of unemployment

Several unit root tests in panel data have recently been proposed. The test developed by Harris and Tzavalis (1999 JoE) performs particularly well when the time dimension is moderate in relation to the cross-section dimension. However, in common with the traditional tests designed for the unidimensional case, it was found to perform poorly when there is a structural break in the time series under the alternative. Here we derive the asymptotic distribution of the test allowing for a shift in the mean, and assess the small sample performance. We apply this new test to show how the hypothesis of (perfect) hysteresis in Spanish unemployment is rejected in favour of the alternative of the natural unemployment rate, when the possibility of a change in the latter is considered.

Yield of intermittent versus continuous EEG in comatose survivors of cardiac arrest treated with hypothermia.

INTRODUCTION: Electroencephalography (EEG) has a central role in the outcome prognostication in subjects with anoxic/hypoxic encephalopathy following a cardiac arrest (CA). Continuous EEG monitoring (cEEG) has been consistently developed and studied; however, its yield as compared to repeated standard EEG (sEEG) is unknown. METHODS: We studied a prospective cohort of comatose adults treated with therapeutic hypothermia (TH) after a CA. cEEG data regarding background activity and epileptiform components were compared to two 20 minute sEEG extracted from the cEEG recording (one during TH, and one in early normothermia). RESULTS: In this cohort, 34 recordings were studied. During TH, the agreement between cEEG and sEEG was 97.1% (95% CI: 84.6 - 99.9%) for background discontinuity and reactivity evaluation, while it was 94.1% (95% CI 80.3 - 99.2%) regarding epileptiform activity. In early normothermia, we did not find any discrepancies. Thus, concordance was very good during TH (kappa 0.83), and optimal during normothermia (kappa=1). The median delay between CA and the first EEG reactivity testing was 18 hours (range: 4.75 - 25) for patients with perfect agreement and 10 hours (range: 5.75 - 10.5) for the three patients in whom there were discordant findings (P=0.02, Wilcoxon). CONCLUSION: Standard intermittent EEG has comparable performance than continuous EEG both for variables important for outcome prognostication (EEG reactivity) and identification of epileptiform transients in this relatively small sample of comatose survivors of CA. This finding has an important practical implication, especially for centers where EEG resources are limited.

Root system distribution of sugar cane as related to nitrogen fertilization, evaluated by two methods: monolith and probes

Few studies on sugar cane have evaluated the root system of the crop, in spite of its importance. This is mainly due to the difficulty of evaluation and high variability of results. The objective of this study was to develop an evaluation method of the cane root system by means of probes so as to evaluate the mass, distribution and metabolically active roots related to N fertilization at planting. For this purpose, an experiment was conducted in an Arenic Kandiustults with medium texture in Jaboticabal/SP, in a randomized block design with four replications and four treatments: control (without N) and 40, 80 and 120 kg ha-1 of N applied in the form of urea in the planting furrow of the cane variety SP81 3250. One week before harvest, a urea-15N solution was applied at the cane stalk base to detect active metabolism in the root system. Trenches of 1.5 m length and 0.6 m depth were opened between two sugar cane rows for root sampling by two methods: monoliths (0.3, 0.2 and 0.15 m wide, deep and long respectively) taken from the trench wall and by probe (internal diameter 0.055 m). For each method, 15 samples per plot were collected. The roots were separated from the soil in a sieve (2 mm mesh), oven-dried (at 65 ºC) and the dry matter was measured. Root sampling by probes resulted in root mass that did not differ from the evaluation in monoliths, indicating that this evaluation method may be used for sugar cane root mass, although neither the root distribution in the soil profile nor the rhizome mass were efficiently evaluated, due to the small sample volume. Nitrogen fertilization at planting did not result in a greater root accumulation in the sugar cane plant, but caused changes in the distribution of the root system in the soil. The absence of N fertilization led to a better root distribution in the soil profile, with 50, 34 and 16 % in the 0-0.2, 0.2-0.4 and 0.4-0.6 m layers, respectively; in the fertilized treatments the roots were concentrated in the surface layer, with on average 70, 17 and 13 % for the same layers. The metabolically active roots were concentrated in the center of the cane stool, amounting to 40 % of the total root mass, regardless of N fertilization (application of 120 kg ha-1 N or without N).

Level shifts in a panel data based unit root test : an application to the rate of unemployment

Several unit root tests in panel data have recently been proposed. The test developed by Harris and Tzavalis (1999 JoE) performs particularly well when the time dimension is moderate in relation to the cross-section dimension. However, in common with the traditional tests designed for the unidimensional case, it was found to perform poorly when there is a structural break in the time series under the alternative. Here we derive the asymptotic distribution of the test allowing for a shift in the mean, and assess the small sample performance. We apply this new test to show how the hypothesis of (perfect) hysteresis in Spanish unemployment is rejected in favour of the alternative of the natural unemployment rate, when the possibility of a change in the latter is considered.

High b-value diffusion-weighted imaging: A sensitive method to reveal white matter differences in schizophrenia.

Over the last 10 years, diffusion-weighted imaging (DWI) has become an important tool to investigate white matter (WM) anomalies in schizophrenia. Despite technological improvement and the exponential use of this technique, discrepancies remain and little is known about optimal parameters to apply for diffusion weighting during image acquisition. Specifically, high b-value diffusion-weighted imaging known to be more sensitive to slow diffusion is not widely used, even though subtle myelin alterations as thought to happen in schizophrenia are likely to affect slow-diffusing protons. Schizophrenia patients and healthy controls were scanned with a high b-value (4000s/mm(2)) protocol. Apparent diffusion coefficient (ADC) measures turned out to be very sensitive in detecting differences between schizophrenia patients and healthy volunteers even in a relatively small sample. We speculate that this is related to the sensitivity of high b-value imaging to the slow-diffusing compartment believed to reflect mainly the intra-axonal and myelin bound water pool. We also compared these results to a low b-value imaging experiment performed on the same population in the same scanning session. Even though the acquisition protocols are not strictly comparable, we noticed important differences in sensitivities in the favor of high b-value imaging, warranting further exploration.

Host genetic determinants of spontaneous hepatitis C clearance

Acute infection with the hepatitis C virus (HCV) induces a wide range of innate and adaptive immune responses. A total of 20-50% of acutely HCV-infected individuals permanently control the virus, referred to as 'spontaneous hepatitis C clearance', while the infection progresses to chronic hepatitis C in the majority of cases. Numerous studies have examined host genetic determinants of hepatitis C infection outcome and revealed the influence of genetic polymorphisms of human leukocyte antigens, killer immunoglobulin-like receptors, chemokines, interleukins and interferon-stimulated genes on spontaneous hepatitis C clearance. However, most genetic associations were not confirmed in independent cohorts, revealed opposing results in diverse populations or were limited by varying definitions of hepatitis C outcomes or small sample size. Coordinated efforts are needed in the search for key genetic determinants of spontaneous hepatitis C clearance that include well-conducted candidate genetic and genome-wide association studies, direct sequencing and follow-up functional studies.