The transform likelihood ratio method for rare event simulation with heavy tails

We present a novel method, called the transform likelihood ratio (TLR) method, for estimation of rare event probabilities with heavy-tailed distributions. Via a simple transformation ( change of variables) technique the TLR method reduces the original rare event probability estimation with heavy tail distributions to an equivalent one with light tail distributions. Once this transformation has been established we estimate the rare event probability via importance sampling, using the classical exponential change of measure or the standard likelihood ratio change of measure. In the latter case the importance sampling distribution is chosen from the same parametric family as the transformed distribution. We estimate the optimal parameter vector of the importance sampling distribution using the cross-entropy method. We prove the polynomial complexity of the TLR method for certain heavy-tailed models and demonstrate numerically its high efficiency for various heavy-tailed models previously thought to be intractable. We also show that the TLR method can be viewed as a universal tool in the sense that not only it provides a unified view for heavy-tailed simulation but also can be efficiently used in simulation with light-tailed distributions. We present extensive simulation results which support the efficiency of the TLR method.

On a resampling approach for tests on the number of clusters with mixture model-based clustering of tissue samples

We consider the problem of assessing the number of clusters in a limited number of tissue samples containing gene expressions for possibly several thousands of genes. It is proposed to use a normal mixture model-based approach to the clustering of the tissue samples. One advantage of this approach is that the question on the number of clusters in the data can be formulated in terms of a test on the smallest number of components in the mixture model compatible with the data. This test can be carried out on the basis of the likelihood ratio test statistic, using resampling to assess its null distribution. The effectiveness of this approach is demonstrated on simulated data and on some microarray datasets, as considered previously in the bioinformatics literature. (C) 2004 Elsevier Inc. All rights reserved.

Selection pressure-driven evolution of the Epstein-Barr virus-encoded oncogene LMP1 in virus isolates from southeast Asia

The geographically constrained distribution of Epstein-Barr virus (EBV)-associated nasopharyngeal carcinoma (NPC) in southeast Asian populations suggests that both viral and host genetics may influence disease risk. Although susceptibility loci have been mapped within the human genome, the role of viral genetics in the focal distribution of NPC remains an enigma. Here we report a molecular phylogenetic analysis of an NPC-associated viral oncogene, LMP1, in a large panel of EBV isolates from southeast Asia and from Papua New Guinea, Africa, and Australia, regions of the world where NPC is and is not endemic, respectively. This analysis revealed that LMP1 sequences show a distinct geographic structure, indicating that the southeast Asian isolates have evolved as a lineage distinct from those of Papua New Guinea, African, and Australian isolates. Furthermore, a likelihood ratio test revealed that the C termini of the LMP1 sequences of the southeast Asian lineage are under significant positive selection pressure, particularly at some sites within the C-terminal activator regions. We also present evidence that although the N terminus and transmembrane region of LMP1 have undergone recombination, the C-terminal region of the gene has evolved without any history of recombination. Based on these observations, we speculate that selection pressure may be driving the LMP1 sequences in virus isolates from southeast Asia towards a more malignant phenotype, thereby influencing the endemic distribution of NPC in this region.

Detecting finite bandwidth periodic signals in stationary noise using the signal coherence spectrum

All signals that appear to be periodic have some sort of variability from period to period regardless of how stable they appear to be in a data plot. A true sinusoidal time series is a deterministic function of time that never changes and thus has zero bandwidth around the sinusoid's frequency. A zero bandwidth is impossible in nature since all signals have some intrinsic variability over time. Deterministic sinusoids are used to model cycles as a mathematical convenience. Hinich [IEEE J. Oceanic Eng. 25 (2) (2000) 256-261] introduced a parametric statistical model, called the randomly modulated periodicity (RMP) that allows one to capture the intrinsic variability of a cycle. As with a deterministic periodic signal the RMP can have a number of harmonics. The likelihood ratio test for this model when the amplitudes and phases are known is given in [M.J. Hinich, Signal Processing 83 (2003) 1349-13521. A method for detecting a RMP whose amplitudes and phases are unknown random process plus a stationary noise process is addressed in this paper. The only assumption on the additive noise is that it has finite dependence and finite moments. Using simulations based on a simple RMP model we show a case where the new method can detect the signal when the signal is not detectable in a standard waterfall spectrograrn display. (c) 2005 Elsevier B.V. All rights reserved.

Trade liberalisation and technical efficiency: Evidence from Bangladesh manufacturing industries

The paper investigates the effects of trade liberalisation on the technical efficiency of the Bangladesh manufacturing sector by estimating a combined stochastic frontier-inefficiency model using panel data for the period 197894 for 25 three-digit level industries. The results show that the overall technical efficiency of the manufacturing sector as well as the technical efficiencies of the majority of the individual industries has increased over time. The findings also clearly suggest that trade liberalisation, proxied by export orientation and capital deepening, has had significant impact on the reduction of the overall technical inefficiency. Similarly, the scale of operation and the proportion of non-production labour in total employment appear as important determinants of technical inefficiency. The evidence also indicates that both export-promoting and import-substituting industries have experienced rises in technical efficiencies over time. Besides, the results are suggestive of neutral technical change, although (at the 5 per cent level of significance) the empirical results indicate that there was no technical change in the manufacturing industries. Finally, the joint test based on the likelihood ratio (LR) test rejects the Cobb-Douglas production technology as description of the database given the specification of the translog production technology.

Association between polymorphisms in the progesterone receptor gene and endometriosis

The progesterone receptor (PR) is a candidate gene for the development of endometriosis, a complex disease with strong hormonal features, common in women of reproductive age. We typed the 306 base pair Alu insertion (AluIns) polymorphism in intron G of PR in 101 individuals, estimated linkage disequilibrium (LD) between five single-nucleotide polymorphisms (SNPs) across the PR locus in 980 Australian triads (endometriosis case and two parents) and used transmission disequilibrium testing (TDT) for association with endometriosis. The five SNPs showed strong pairwise LD, and the AluIns was highly correlated with proximal SNPs rs1042839 ({Delta}2 = 0.877, D9 = 1.00, P < 0.0001) and rs500760 ({Delta}2 = 0.438, D9 = 0.942, P < 0.0001). TDT showed weak evidence of allelic association between endometriosis and rs500760 (P = 0.027) but not in the expected direction. We identified a common susceptibility haplotype GGGCA across the five SNPs (P = 0.0167) in the whole sample, but likelihood ratio testing of haplotype transmission and non-transmission of the AluIns and flanking SNPs showed no significant pattern. Further, analysis of our results pooled with those from two previous studies suggested that neither the T2 allele of the AluIns nor the T1/T2 genotype was associated with endometriosis.

Is pulse palpation helpful in detecting atrial fibrillation? A systematic review

Background Atrial fibrillation in the elderly is common and potentially life threatening. The classical sign of atrial fibrillation is an irregularly irregular pulse. Objective The objective of this research was to determine the accuracy of pulse palpation to detect atrial fibrillation. Methods We searched Medline, EMBASE, and the reference lists of review articles for studies that compared pulse palpation with the electrocardiogram (ECG) diagnosis of atrial fibrillation. Two reviewers independently assessed the search results to determine the eligibility of studies, extracted data, and assessed the quality of the studies. Results We identified 3 studies (2385 patients) that compared pulse palpation with ECG. The estimated sensitivity of pulse palpation ranged from 91% to 100%, while specificity ranged from 70% to 77%. Pooled sensitivity was 94% (95% confidence interval [CI], 84%-97%) and pooled specificity was 72% (95% CI 69%-75%). The pooled positive likelihood ratio was 3.39, while the pooled negative likelihood ratio was 0.10. Conclusions Pulse palpation has a high sensitivity but relatively low specificity for atrial fibrillation. It is therefore useful for ruling out atrial fibrillation. It may also be a useful screen to apply opportunistically for previously undetected atrial fibrillation. Assuming a prevalence of 3% for undetected atrial fibrillation in patients older than 65 years, and given the test's sensitivity and specificity, opportunistic pulse palpation in this age group would detect an irregular pulse in 30% of screened patients, requiring further testing with ECG. Among screened patients, 0.2% would have atrial fibrillation undetected with pulse palpation.

Use of umbilical artery Doppler velocimetry in the monitoring of pregnancy in women with pre-existing diabetes

Background: The usefulness of umbilical artery Doppler velocimetry for the monitoring of diabetic pregnancies is controversial. The aim of the present study was to assess whether umbilical artery Doppler velocity waveform analysis can predict adverse perinatal outcomes for pregnancies complicated by pre-existing diabetes mellitus. Methods: All diabetic pregnancies (type 1 and 2) delivered at Mater Mothers' Hospital, Queensland, between 1 January 1995 and 31 December 1999 were included. All pregnant diabetic women were monitored with umbilical artery Doppler velocimetry at 28, 32, 36, and 38 weeks' gestation. Adverse perinatal outcome was defined as pregnancies with one or more of the following: small-for-gestational age, Caesarean section for non-reassuring cardiotocography, fetal acidaemia at delivery, 1-min Apgar of 3 or less, 5-min Apgar of less than 7, hypoxic ischaemic encephalopathy or perinatal death. Abnormal umbilical artery Doppler velocimetry was defined as a pulsatility index of 95th centile or higher for gestation. Results: One hundred and four pregnancies in women with pre-existing diabetes had umbilical arterial Doppler studies carried out during the study period. Twenty-three pregnancies (22.1%) had an elevated pulsatility index. If the scans were carried out within 2 weeks of delivery, 71% of pregnancies with abnormal umbilical Doppler had adverse outcomes (P < 0.01; likelihood ratio, 4.2). However, the sensitivity was 35%; specificity was 94%; positive predictive value was 80%; and negative predictive value was 68%. Only 30% of women with adverse perinatal outcomes had abnormal umbilical arterial Doppler flow. Conclusion: Umbilical artery Doppler velocimetry is not a good predictor of adverse perinatal outcomes in diabetic pregnancies.

Genomewide Linkage Scan of 409 European-ancestry and African American Families with Schizophrenia: Suggestive Evidence of Linkage at 8p23.3-p21.2 and 11p13.1-q14.1 in the Combined Sample

We report the clinical characteristics of a schizophrenia sample of 409 pedigrees-263 of European ancestry ( EA) and 146 of African American ancestry ( AA)-together with the results of a genome scan ( with a simple tandem repeat polymorphism interval of 9 cM) and follow-up fine mapping. A family was required to have a proband with schizophrenia ( SZ) and one or more siblings of the proband with SZ or schizoaffective disorder. Linkage analyses included 403 independent full-sibling affected sibling pairs ( ASPs) ( 279 EA and 124 AA) and 100 all-possible half-sibling ASPs ( 15 EA and 85 AA). Nonparametric multipoint linkage analysis of all families detected two regions with suggestive evidence of linkage at 8p23.3-q12 and 11p11.2-q22.3 ( empirical Z likelihood-ratio score [ Z(lr)] threshold >= 2.65) and, in exploratory analyses, two other regions at 4p16.1-p15.32 in AA families and at 5p14.3-q11.2 in EA families. The most significant linkage peak was in chromosome 8p; its signal was mainly driven by the EA families. Z(lr) scores >= 2.0 in 8p were observed from 30.7 cM to 61.7 cM ( Center for Inherited Disease Research map locations). The maximum evidence in the full sample was a multipoint Z(lr) of 3.25 ( equivalent Kong-Cox LOD of 2.30) near D8S1771 ( at 52 cM); there appeared to be two peaks, both telomeric to neuregulin 1 ( NRG1). There is a paracentric inversion common in EA individuals within this region, the effect of which on the linkage evidence remains unknown in this and in other previously analyzed samples. Fine mapping of 8p did not significantly alter the significance or length of the peak. We also performed fine mapping of 4p16.3-p15.2, 5p15.2-q13.3, 10p15.3-p14, 10q25.3-q26.3, and 11p13-q23.3. The highest increase in Z(lr) scores was observed for 5p14.1-q12.1, where the maximum Z(lr) increased from 2.77 initially to 3.80 after fine mapping in the EA families.

Structural change in macroeconomic time series: A complex systems perspective

We demonstrate that the process of generating smooth transitions Call be viewed as a natural result of the filtering operations implied in the generation of discrete-time series observations from the sampling of data from an underlying continuous time process that has undergone a process of structural change. In order to focus discussion, we utilize the problem of estimating the location of abrupt shifts in some simple time series models. This approach will permit its to address salient issues relating to distortions induced by the inherent aggregation associated with discrete-time sampling of continuous time processes experiencing structural change, We also address the issue of how time irreversible structures may be generated within the smooth transition processes. (c) 2005 Elsevier Inc. All rights reserved.

Machine learning for matching astronomy catalogues

An emerging issue in the field of astronomy is the integration, management and utilization of databases from around the world to facilitate scientific discovery. In this paper, we investigate application of the machine learning techniques of support vector machines and neural networks to the problem of amalgamating catalogues of galaxies as objects from two disparate data sources: radio and optical. Formulating this as a classification problem presents several challenges, including dealing with a highly unbalanced data set. Unlike the conventional approach to the problem (which is based on a likelihood ratio) machine learning does not require density estimation and is shown here to provide a significant improvement in performance. We also report some experiments that explore the importance of the radio and optical data features for the matching problem.

Partitioned likelihood support and the evaluation of data set conflict

In simultaneous analyses of multiple data partitions, the trees relevant when measuring support for a clade are the optimal tree, and the best tree lacking the clade (i.e., the most reasonable alternative). The parsimony-based method of partitioned branch support (PBS) forces each data set to arbitrate between the two relevant trees. This value is the amount each data set contributes to clade support in the combined analysis, and can be very different to support apparent in separate analyses. The approach used in PBS can also be employed in likelihood: a simultaneous analysis of all data retrieves the maximum likelihood tree, and the best tree without the clade of interest is also found. Each data set is fitted to the two trees and the log-likelihood difference calculated, giving partitioned likelihood support (PLS) for each data set. These calculations can be performed regardless of the complexity of the ML model adopted. The significance of PLS can be evaluated using a variety of resampling methods, such as the Kishino-Hasegawa test, the Shimodiara-Hasegawa test, or likelihood weights, although the appropriateness and assumptions of these tests remains debated.

A major quantitative trait locus for CD4-CD8 ratio is located on chromosome 11

CD4-CD8 ratio is an important diagnostic measure of immune system functioning. In particular, CD4-CD8 ratio predicts the time taken for progression of HIV infection to acquired immune deficiency syndrome (AIDS) and the long-term survival of AIDS patients. To map genes that regulate differences between healthy individuals in CD4-CD8 ratio, we typed 757 highly polymorphic microsatellite markers at an average spacing of similar to5 cM across the genome in 405 pairs of dizygotic twins at ages 12, 14 and 16. We used multipoint variance components linkage analysis to test for linkage between marker loci and CD4-CD8 ratio at each age. We found suggestive evidence of linkage on chromosome 11p in 12-year-old twins (LOD=2.55, P=0.00031) and even stronger evidence of linkage in the same region at age 14 (LOD 3.51, P=0.00003). Possible candidate genes include CD5 and CD6, which encode cell membrane proteins involved in the positive selection of thymocytes. We also found suggestive evidence of linkage at other areas of the genome including regions on chromosomes 1, 3, 4, 5, 6, 12, 13, 15, 17 and 22.

The aldosterone-renin ratio in screening for primary aldosteronism

Recognition that primary aldosteronism (PAL) is a common specifically treatable form of hypertension and that most patients are normokalemic has led to a marked increase in demand for aldosterone/renin ratio (ARR) testing as a means of screening for this disorder. The value of this screening test depends on an appreciation of many factors (such as diet, posture, time of day, presence of hypokalemia, medications, age, and renal function), which can affect the results, on the care with which these factors are either controlled or their effects taken into account, and on access to reliable and reproducible assays for renin and aldosterone. Even then, physiological day-to-day variability reduces the value of a single estimation, and repeated testing is necessary before a decision that PAL is highly likely (warranting further testing) or highly unlikely can be made. Provided that testing of aldosterone suppressibility is always carried out to confirm or exclude the diagnosis, and the subtype is determined by hybrid gene testing and adrenal venous sampling, wide application of the ARR can have a major beneficial clinical impact with improved therapeutic outcomes, including possible cure in those with unilateral disease.

Spatio-temporal scanning and statistical test of the accelerating moment release (AMR) model using Australian earthquake data

The Accelerating Moment Release (AMR) preceding earthquakes with magnitude above 5 in Australia that occurred during the last 20 years was analyzed to test the Critical Point Hypothesis. Twelve earthquakes in the catalog were chosen based on a criterion for the number of nearby events. Results show that seven sequences with numerous events recorded leading up to the main earthquake exhibited accelerating moment release. Two occurred near in time and space to other earthquakes preceded by AM R. The remaining three sequences had very few events in the catalog so the lack of AMR detected in the analysis may be related to catalog incompleteness. Spatio-temporal scanning of AMR parameters shows that 80% of the areas in which AMR occurred experienced large events. In areas of similar background seismicity with no large events, 10 out of 12 cases exhibit no AMR, and two others are false alarms where AMR was observed but no large event followed. The relationship between AMR and Load-Unload Response Ratio (LURR) was studied. Both methods predict similar critical region sizes, however, the critical point time using AMR is slightly earlier than the time of the critical point LURR anomaly.