5 resultados para Hypothesis test
em Biblioteca Digital da Produção Intelectual da Universidade de São Paulo
Resumo:
The distribution of delta ferrite fraction was measured with the magnetic method in specimens of different stainless steel compositions cast by the investment casting (lost wax) process. Ferrite fraction measurements published in the literature for stainless steel cast samples were added to the present work data, enabling an extensive analysis about practical methods to calculate delta ferrite fractions in stainless steel castings. Nineteen different versions of practical methods were formed using Schaeffler, DeLong, and Siewert diagrams and the nickel and chromium equivalent indexes suggested by several authors. These methods were evaluated by a detailed statistical analysis, showing that the Siewert diagram, including its equivalent indexes and iso-ferrite lines, gives the lowest relative errors between calculated and measured delta ferrite fractions. Although originally created for stainless steel welds, this diagram gives relative errors lower than those for the current ASTM standard method (800/A 800M-01), developed to predict ferrite fractions in stainless steel castings. Practical methods originated from a combination of different chromium/nickel equivalent indexes and the iso-ferrite lines from Schaeffler diagram give the lowest relative errors when compared with combinations using other iso-ferrite line diagrams. For the samples cast in the present work, an increase in cooling rate from 0.78 to 2.7 K/s caused a decrease in the delta ferrite fraction, but a statistical hypothesis test revealed that this effect is significant in only 50% of the samples that have ferrite in their microstructures.
Resumo:
This paper aims to discuss and test the hypothesis raised by Fusar-Poli [Fusar-Poli P. Can neuroimaging prove that schizophrenia is a brain disease? A radical hypothesis. Medical Hypotheses in press, corrected proof] that ""on the basis of the available imaging literature there is no consistent evidence to reject the radical and provocative hypothesis that schizophrenia is not a brain disease"". To achieve this goal, all meta-analyses on `fMRI and schizophrenia` published during the current decade and indexed in Pubmed were summarized, as much as some other useful information, e.g., meta-analyses on genetic risk factors. Our main conclusion is that the literature fully supports the hypothesis that schizophrenia is a syndrome (not a disease) associated with brain abnormalities, despite the fact that there is no singular and reductionist pathway from the nosographic entity (schizophrenia) to its causes. This irreducibility is due to the fact that the syndrome has more than one dimension (e.g., cognitive, psychotic and negative) and each of them is related to abnormalities in specific neuronal networks. A psychiatric diagnosis is a statistical procedure; these dimensions are not identically represented in each diagnosticated case and this explains the existence of more than one pattern of brain abnormalities related to schizophrenia. For example, chronification is associated with negativism while the first psychotic episode is not; in that sense, the same person living with schizophrenia may reveal different symptoms and fMRI patterns along the course of his life, and this is precisely what defines schizophrenia since the time when it was called Dementia Praecox (first by pick then by Kraepelin). It is notable that 100% of the collected meta-analyses on `fMRI and schizophrenia` reveal positive findings. Moreover, all meta-analyses that found positive associations between schizophrenia and genetic risk factors have to do with genes (SNPs) especially activated in neuronal tissue of the central nervous system (CNS), suggesting that, to the extent these polymorphisms are related to schizophrenia`s etiology, they are also related to abnormal brain activity. (C) 2009 Elsevier Ltd. All rights reserved.
Resumo:
Objectives. To verify the hypothesis that crack analysis and a mechanical test would rank a series of composites in a similar order with respect to polymerization stress. Also, both tests would show similar relationships between stress and composite elastic modulus and/or shrinkage. Methods. Soda-lime glass discs (2-mm thick) with a central perforation (3.5-mm diameter) received four Vickers indentations 500 mu m from the cavity margin. The indent cracks were measured (500x) prior and 10 min after the cavity was restored with one of six materials (Kalore/KL, Gradia/GR, Ice/IC, Wave/WV, Majesty Flow/MF, and Majesty Posterior/MP). Stresses at the indent site were calculated based on glass fracture toughness and increase in crack length. Stress at the bonded interface was calculated using the equation for an internally pressurized cylinder. The mechanical test used a universal testing machine and glass rods (5-mm diameter) as substrate. An extensometer monitored specimen height (2 mm). Nominal stress was calculated dividing the maximum shrinkage force by the specimen cross-sectional area. Composite elastic modulus was determined by nanoindentation and post-gel shrinkage was measured using strain gages. Data were subjected to one-way ANOVA/Tukey or Kruskal-Wallis/Mann-Whitney tests (alpha: 5%). Results. Both tests grouped the composites in three statistical subsets, with small differences in overlapping between the intermediate subset (MF, WV) and the highest (MP, IC) or the lowest stress materials (KL, GR). Higher stresses were developed by composites with high modulus and/or high shrinkage. Significance. Crack analysis demonstrated to be as effective as the mechanical test to rank composites regarding polymerization stress. (c) 2012 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
Resumo:
To estimate causal relationships, time series econometricians must be aware of spurious correlation, a problem first mentioned by Yule (1926). To deal with this problem, one can work either with differenced series or multivariate models: VAR (VEC or VECM) models. These models usually include at least one cointegration relation. Although the Bayesian literature on VAR/VEC is quite advanced, Bauwens et al. (1999) highlighted that "the topic of selecting the cointegrating rank has not yet given very useful and convincing results". The present article applies the Full Bayesian Significance Test (FBST), especially designed to deal with sharp hypotheses, to cointegration rank selection tests in VECM time series models. It shows the FBST implementation using both simulated and available (in the literature) data sets. As illustration, standard non informative priors are used.
Resumo:
The asymptotic expansion of the distribution of the gradient test statistic is derived for a composite hypothesis under a sequence of Pitman alternative hypotheses converging to the null hypothesis at rate n(-1/2), n being the sample size. Comparisons of the local powers of the gradient, likelihood ratio, Wald and score tests reveal no uniform superiority property. The power performance of all four criteria in one-parameter exponential family is examined.