Using artificial neural networks to generate trading signals for crude oil, copper and gold futures

In this thesis, a feed-forward, back-propagating Artificial Neural Network using the gradient descent algorithm is developed to forecast the directional movement of daily returns for WTI, gold and copper futures. Out-of-sample back-test results vary, with some predictive abilities for copper futures but none for either WTI or gold. The best statistically significant hit rate achieved was 57% for copper with an absolute return Sharpe Ratio of 1.25 and a benchmarked Information Ratio of 2.11.

Growth rate of a terra firme rain forest in Brazilian Amazonia over an eight-year period in response to logging

This paper deals with growth rates of trees > 5cm dbh over an eight-year period from 257 species at the Tapajós National Forest. The discussion is centred on the behaviour of the forest after logging. Permanent sample plots were established in 1981 and measured at the first time. The area was logged in 1982. Measurements after logging occurred in 1983, 1987 and 1989. Considering all species together, diameter increment was similar for both intensities of logging until five years after logging. Light-demanding species showed significantly higher growth rates than shade-tolerant species in the logged forest, with greater increment in the heavier treatment intensity. Commercial species also had higher growth rates in the heavier logged area, although those were significantly different only in the period from one to five years after logging. In the undisturbed forest, growth rates increased with increasing dbh size. At species level, growth rate varied between and within treatments, as well as between trees within species, depending mainly on degree of canopy opening. The logging favoured the growth of commercial species, chiefly the light-demanders. Therefore, if the same growth conditions continue being given, for example by silvicultural treatments, to those species of commercial interest, the forest would reach a stock available for harvesting around year 30 after logging. However, the high variation in increment rates indicates that an eight-year period is not sufficient to allow predictions on cutting cycles or polycyclic management systems for the study forest.

High strain rate constitutive modeling forhistorical structures subjected to blast loading

Doctoral Thesis Civil Engineering

The short and long-run determinants of the real exchange rate in Mexico

This paper explores the real exchange rate behavior in Mexico from 1960 until 2005. Since the empirical analysis reveals that the real exchange rate is not mean reverting, we propose that economic fundamental variables affect its evolution in the long-run. Therefore, based on equilibrium exchange rate paradigms, we propose a simple model of real exchange rate determination which includes the relative labor productivity, the real interest rates and the net foreign assets over a long period of time. Our analysis also considers the dynamic adjustment in response to shocks through impulse response functions derived from the multivariate VAR model.

Evolutionary implications of a high selfing rate in the freshwater snail Lymnaea truncatula.

Self-compatible hermaphroditic organisms that mix self-fertilization and outcrossing are of great interest for investigating the evolution of mating systems. We investigate the evolution of selfing in Lymnaea truncatula, a self-compatible hermaphroditic freshwater snail. We first analyze the consequences of selfing in terms of genetic variability within and among populations and then investigate how these consequences along with the species ecology (harshness of the habitat and parasitism) might govern the evolution of selfing. Snails from 13 localities (classified as temporary or permanent depending on their water availability) were sampled in western Switzerland and genotyped for seven microsatellite loci. F(IS) (estimated on adults) and progeny array analyses (on hatchlings) provided similar selfing rate estimates of 80%. Populations presented a low polymorphism and were highly differentiated (F(ST) = 0.58). Although the reproductive assurance hypothesis would predict higher selfing rate in temporary populations, no difference in selfing level was observed between temporary and permanent populations. However, allelic richness and gene diversity declined in temporary habitats, presumably reflecting drift. Infection levels varied but were not simply related to either estimated population selfing rate or to differences in heterozygosity. These findings and the similar selfing rates estimated for hatchlings and adults suggest that within-population inbreeding depression is low in L. truncatula.

Rate and predictors of service disengagement in an epidemiological first-episode psychosis cohort.

OBJECTIVES: To assess the prevalence and predictors of service disengagement in a treated epidemiological cohort of first-episode psychosis (FEP) patients. METHODS: The Early Psychosis Prevention and Intervention Centre (EPPIC) in Australia admitted 786 FEP patients from January 1998 to December 2000. Treatment at EPPIC is scheduled for 18 months. Data were collected from patients' files using a standardized questionnaire. Seven hundred four files were available; 44 were excluded, because of a non-psychotic diagnosis at endpoint (n=43) or missing data on service disengagement (n=1). Rate of service disengagement was the outcome of interest, as well as pre-treatment, baseline, and treatment predictors of service disengagement, which were examined via Cox proportional hazards models. RESULTS: 154 patients (23.3%) disengaged from service. A past forensic history (Hazard ratio [HR]=1.69; 95%CI 1.17-2.45), lower severity of illness at baseline (HR=0.59; 95%CI 0.48-0.72), living without family at discharge (HR=1.75; 95%CI 1.22-2.50) and persistence of substance use disorder during treatment (HR=2.30; 95%CI 1.45-3.66) were significant predictors of disengagement from service. CONCLUSIONS: While engagement strategies are a core element in the treatment of first-episode psychosis, particular attention should be paid to these factors associated with disengagement. Involvement of the family in the treatment process, and focusing on reduction of substance use, need to be pursued in early intervention services.

Estimation de la fonction rénale par l'equation MDRD: intérêt et limites pour l'adaptation des doses de médicaments [Renal function estimation by MDRD equation: interest and limitations for drug dosing].

The MDRD (Modification of diet in renal disease) equation enables glomerular filtration rate (GFR) estimation from serum creatinine only. Thus, the laboratory can report an estimated GFR (eGFR) with each serum creatinine assessment, increasing therefore the recognition of renal failure. Predictive performance of MDRD equation is better for GFR < 60 ml/min/1,73 m2. A normal or near-normal renal function is often underestimated by this equation. Overall, MDRD provides more reliable estimations of renal function than the Cockcroft-Gault (C-G) formula, but both lack precision. MDRD is not superior to C-G for drug dosing. Being adjusted to 1,73 m2, MDRD eGFR has to be back adjusted to the patient's body surface area for drug dosing. Besides, C-G has the advantage of a greater simplicity and a longer use.

Increasing lactate turnover rate during exercise by means of altitude training and fructose ingestion : effects on substrate metabolism and endurance performance

Summary : With regard to exercise metabolism, lactate was long considered as a dead-end waste product responsible for muscle fatigue and a limiting factor for motor performance. However, a large body of evidence clearly indicates that lactate is an energy efficient metabolite able to link the glycolytic pathway with aerobic metabolism and has endocrine-like actions, rather than to be a dead-end waste product. Lactate metabolism is also known to be quickly upregulated by regular endurance training and is thought to be related to exercise performance. However, to what extent its modulation can increase exercise performance in already endurance-trained subjects is unknown. The general hypothesis of this work was therefore that increasing either lactate metabolic clearance rate or lactate availability could, in turn, increase endurance performance. The first study (Study I) aimed at increasing the lactate clearance rate by means of assumed interaction effects of endurance training and hypoxia on lactate metabolism and endurance performance. Although this study did not demonstrate any interaction of training and hypoxia on both lactate metabolism and endurance performance, a significant deleterious effect of endurance training in hypoxia was shown on glucose homeostasis. The methods used to determine lactate kinetics during exercise exhibited some limitations, and the second study did delineate some of the issues raised (Study 2). The third study (Study 3) investigated the metabolic and performance effects of increasing plasma lactate production and availability during prolonged exercise in the fed state. A nutritional intervention was used for this purpose: part of glucose feedings ingested during the control condition was substituted by fructose. The results of this study showed a significant increase of lactate turnover rate, quantified the metabolic fate of fructose; and demonstrated a significant decrease of lipid oxidation and glycogen breakdown. In contrast, endurance performance appeared to be unmodified by this dietary intervention, being at odds with recent reports. Altogether the results of this thesis suggest that in endurance athletes the relationship between endurance performance and lactate turnover rate remains unclear. Nonetheless, the result of the present study raises questions and opens perspectives on the rationale of using hypoxia as a therapeutic aid for the treatment of insulin resistance. Moreover, the results of the second study open perspectives on the role of lactate as an intermediate metabolite and its modulatory effects on substrate metabolism during exercise. Additionally it is suggested that the simple nutritional intervention used in the third study can be of interest in the investigation on the aforementioned roles of lactate. Résumé : Lorsque le lactate est évoqué en rapport avec l'exercice, il est souvent considéré comme un déchet métabolique responsable de l'acidose métabolique, de la fatigue musculaire ou encore comme un facteur limitant de la performance. Or la littérature montre clairement que le lactate se révèle être plutôt un métabolite utilisé efficacement par de nombreux tissus par les voies oxydatives et, ainsi, il peut être considéré comme un lien entre le métabolisme glycolytique et le métabolisme oxydatif. De plus on lui prête des propriétés endocrines. Il est connu que l'entraînement d'endurance accroît rapidement le métabolisme du lactate, et il est suggéré que la performance d'endurance est liée à son métabolisme. Toutefois la relation entre le taux de renouvellement du lactate et la performance d'endurance est peu claire, et, de même, de quelle manière la modulation de son métabolisme peut influencer cette dernière. Le but de cette thèse était en conséquence d'investiguer de quelle manière et à quel degré l'augmentation du métabolisme du lactate, par l'augmentation de sa clearance et de son turnover, pouvait à son tour améliorer la performance d'endurance de sujets entraînés. L'objectif de la première étude a été d'augmenter la clearance du lactate par le biais d'un entraînement en conditions hypoxiques chez des cyclistes d'endurance. Basé sur la littérature scientifique existante, on a fait l'hypothèse que l'entraînement d'endurance et l'hypoxie exerceraient un effet synergétique sur le métabolisme du lactate et sur la performance, ce qui permettrait de montrer des relations entre performance et métabolisme du lactate. Les résultats de cette étude n'ont montré aucun effet synergique sur la performance ou le métabolisme du lactate. Toutefois, un effet délétère sur le métabolisme du glucose a été démontré. Quelques limitations de la méthode employée pour la mesure du métabolisme du lactate ont été soulevées, et partiellement résolues dans la seconde étude de ce travail, qui avait pour but d'évaluer la sensibilité du modèle pharmacodynamique utilisé pour le calcul du turnover du lactate. La troisième étude a investigué l'effet d'une augmentation de la lactatémie sur le métabolisme des substrats et sur la performance par une intervention nutritionnelle substituant une partie de glucose ingéré pendant l'exercice par du fructose. Les résultats montrent que les composants dynamiques du métabolisme du lactate sont significativement augmentés en présence de fructose, et que les oxydations de graisse et de glycogène sont significativement diminuées. Toutefois aucun effet sur la performance n'a été démontré. Les résultats de ces études montrent que la relation entre le métabolisme du lactate et la performance reste peu claire. Les résultats délétères de la première étude laissent envisager des pistes de travail, étant donné que l'entraînement en hypoxie est considéré comme outil thérapeutique dans le traitement de pathologies liées à la résistance à l'insuline. De plus les résultats de la troisième étude ouvrent des perspectives de travail quant au rôle du lactate comme intermédiaire métabolique durant l'exercice ainsi que sur ses effets directs sur le métabolisme. Ils suggèrent de plus que la manipulation nutritionnelle simple qui a été utilisée se révèle être un outil prometteur dans l'étude des rôles et effets métaboliques que peut revêtir le lactate durant l'exercice.

Structural macro factors and the affine term structure of interest rates

This work consists of three essays investigating the ability of structural macroeconomic models to price zero coupon U.S. government bonds. 1. A small scale 3 factor DSGE model implying constant term premium is able to provide reasonable a fit for the term structure only at the expense of the persistence parameters of the structural shocks. The test of the structural model against one that has constant but unrestricted prices of risk parameters shows that the exogenous prices of risk-model is only weakly preferred. We provide an MLE based variance-covariance matrix of the Metropolis Proposal Density that improves convergence speeds in MCMC chains. 2. Affine in observable macro-variables, prices of risk specification is excessively flexible and provides term-structure fit without significantly altering the structural parameters. The exogenous component of the SDF is separating the macro part of the model from the term structure and the good term structure fit has as a driving force an extremely volatile SDF and an implied average short rate that is inexplicable. We conclude that the no arbitrage restrictions do not suffice to temper the SDF, thus there is need for more restrictions. We introduce a penalty-function methodology that proves useful in showing that affine prices of risk specifications are able to reconcile stable macro-dynamics with good term structure fit and a plausible SDF. 3. The level factor is reproduced most importantly by the preference shock to which it is strongly and positively related but technology and monetary shocks, with negative loadings, are also contributing to its replication. The slope factor is only related to the monetary policy shocks and it is poorly explained. We find that there are gains in in- and out-of-sample forecast of consumption and inflation if term structure information is used in a time varying hybrid prices of risk setting. In-sample yield forecast are better in models with non-stationary shocks for the period 1982-1988. After this period, time varying market price of risk models provide better in-sample forecasts. For the period 2005-2008, out of sample forecast of consumption and inflation are better if term structure information is incorporated in the DSGE model but yields are better forecasted by a pure macro DSGE model.

Enhancement of viability of radiosensitive (PBMC) and resistant (MDA-MB-231) clones in low-dose-rate cobalt-60 radiation therapy

AbstractObjective:In the present study, the authors investigated the in vitrobehavior of radio-resistant breast adenocarcinoma (MDA-MB-231) cells line and radiosensitive peripheral blood mononuclear cells (PBMC), as a function of different radiation doses, dose rates and postirradiation time kinetics, with a view to the interest of clinical radiotherapy.Materials and Methods:The cells were irradiated with Co-60, at 2 and 10 Gy and two different exposure rates, 339.56 cGy.min–1 and the other corresponding to one fourth of the standard dose rates, present over a 10-year period of cobalt therapy. Post-irradiation sampling was performed at pre-established kinetics of 24, 48 and 72 hours. The optical density response in viability assay was evaluated and a morphological analysis was performed.Results:Radiosensitive PBMC showed decrease in viability at 2 Gy, and a more significant decrease at 10 Gy for both dose rates. MDAMB- 231 cells presented viability decrease only at higher dose and dose rate. The results showed MDA-MB-231 clone expansion at low dose rate after 48–72 hours post-radiation.Conclusion:Low dose rate shows a possible potential clinical impact involving decrease in management of radio-resistant and radiosensitive tumor cell lines in cobalt therapy for breast cancer.

Ecological and Epidemiological Analyses of Multiple Sclerosis Relapse Rate

The underlying cause of many human autoimmune diseases is unknown, but several environmental factors are implicated in triggering the self-destructive immune reactions. Multiple Sclerosis (MS) is a chronic autoimmune disease of the central nervous system, potentially leading to persistent neurological deterioration. The cause of MS is not known, and apart from immunomodulatory treatments there is no cure. In the early phase of the disease, relapsing-remitting MS (RR-MS) is characterized by unpredictable exacerbations of the neurological symptoms called relapses, which can occur at different intervals ranging from 4 weeks to several years. Microbial infections are known to be able to trigger MS relapses, and the patients are instructed to avoid all factors that might increase the risk of infections and to properly use antibiotics as well as to take care of dental hygiene. Among those environmental factors which are known to increase susceptibility to infections, high ambient air inhalable particulate matter levels affect all people within a geographical region. During the period of interest in this thesis, the occurrence of MS relapses could be effectively reduced by injections of interferon, which has immunomodulatory and antiviral properties. In this thesis, ecological and epidemiological analyses were used to study the possible connection between MS relapse occurrence, population level viral infections and air quality factors, as well as the effects of interferon medication. Hospital archive data were collected retrospectively from 1986-2001, a period in time ranging from when interferon medication first became available until just before other disease-modifying MS therapies arrived on the market. The grouped data were studied with logistic regression and intervention analysis, and individual patient data with survival analysis. Interferons proved to be effective in the treatment of MS in this observational study, as the amount of MS exacerbations was lower during interferon use as compared to the time before interferon treatment. A statistically significant temporal relationship between MS relapses and inhalable particular matter (PM₁₀) concentrations was found in this study, which implies that MS patients are affected by the exposure to PM₁₀. Interferon probably protected against the effect of PM₁₀, because a significant increase in the risk of exacerbations was only observed in MS patients without interferon medication following environmental exposure to population level specific viral infections and PM₁₀. Apart from being antiviral, interferon could thus also attenuate the enhancement of immune reactions caused by ambient air PM₁₀. The retrospective approach utilizing carefully constructed hospital records proved to be an economical and reliable source of MS disease information for statistical analyses.

A study on rate correlations of gasification reactions

Gasification offers an environmentally friendly alternative for conventional combustion enabling the use of low grade and troublesome fuel such as municipal waste. While combustion converts fuel directly into thermal energy and noxious gases, gasification thermally converts fuel into gas that can be used in multiple applications. The purpose of this work is to get to know the gasification as a phenomenon and examine the kinetics of gasification. The main interest is in the reaction rates of the most important gasification reactions - water-gas, Boudouard and shift reaction. Reaction rate correlations found in the scientific articles are examined in atmospheric pressure in different temperatures.

Testing Futures Market Efficiency and Optimal Hedge Ratio Estimation: Evidence from Futures and Options on RTS

This study investigates futures market efficiency and optimal hedge ratio estimation. First, cointegration between spot and futures prices is studied using Johansen method, with two different model specifications. If prices are found cointegrated, restrictions on cointegrating vector and adjustment coefficients are imposed, to account for unbiasedness, weak exogeneity and prediction hypothesis. Second, optimal hedge ratios are estimated using static OLS, and time-varying DVEC and CCC models. In-sample and out-of-sample results for one, two and five period ahead are reported. The futures used in thesis are RTS index, EUR/RUB exchange rate and Brent oil, traded in Futures and options on RTS.(FORTS) For in-sample period, data points were acquired from start of trading of each futures contract, RTS index from August 2005, EUR/RUB exchange rate March 2009 and Brent oil October 2008, lasting till end of May 2011. Out-of-sample period covers start of June 2011, till end of December 2011. Our results indicate that all three asset pairs, spot and futures, are cointegrated. We found RTS index futures to be unbiased predictor of spot price, mixed evidence for exchange rate, and for Brent oil futures unbiasedness was not supported. Weak exogeneity results for all pairs indicated spot price to lead in price discovery process. Prediction hypothesis, unbiasedness and weak exogeneity of futures, was rejected for all asset pairs. Variance reduction results varied between assets, in-sample in range of 40-85 percent and out-of sample in range of 40-96 percent. Differences between models were found small, except for Brent oil in which OLS clearly dominated. Out-of-sample results indicated exceptionally high variance reduction for RTS index, approximately 95 percent.

Understanding the in vitro dissolution rate of glasses with respect to future clinical applications

Glass is a unique material with a long history. Several glass products are used daily in our everyday life, often unnoticed. Glass can be found not only in obvious applications such as tableware, windows, and light bulbs, but also in tennis rackets, windmill turbine blades, optical devices, and medical implants. The glasses used at present as implants are inorganic silica-based melt-derived compositions mainly for hard-tissue repair as bone graft substitute in dentistry and orthopedics. The degree of glass reactivity desired varies according to implantation situation and it is vital that the ion release from any glasses used in medical applications is controlled. Understanding the in vitro dissolution rate of glasses provides a first approximation of their behavior in vivo. Specific studies concerning dissolution properties of bioactive glasses have been relatively scarce and mostly concentrated to static condition studies. The motivation behind this work was to develop a simple and accurate method for quantifying the in vitro dissolution rate of highly different types of glass compositions with interest for future clinical applications. By combining information from various experimental conditions, a better knowledge of glass dissolution and the suitability of different glasses for different medical applications can be obtained. Thus, two traditional and one novel approach were utilized in this thesis to study glass dissolution. The chemical durability of silicate glasses was tested in water and TRIS-buffered solution at static and dynamic conditions. The traditional in vitro testing with a TRISbuffered solution under static conditions works well with bioactive or with readily dissolving glasses, and it is easy to follow the ion dissolution reactions. However, in the buffered solution no marked differences between the more durable glasses were observed. The hydrolytic resistance of the glasses was studied using the standard procedure ISO 719. The relative scale given by the standard failed to provide any relevant information when bioactive glasses were studied. However, the clear differences in the hydrolytic resistance values imply that the method could be used as a rapid test to get an overall idea of the biodegradability of glasses. The standard method combined with the ion concentration and pH measurements gives a better estimate of the hydrolytic resistance because of the high silicon amount released from a glass. A sensitive on-line analysis method utilizing inductively coupled plasma optical emission spectrometer and a flow-through micro-volume pH electrode was developed to study the initial dissolution of biocompatible glasses. This approach was found suitable for compositions within a large range of chemical durability. With this approach, the initial dissolution of all ions could be measured simultaneously and quantitatively, which gave a good overall idea of the initial dissolution rates for the individual ions and the dissolution mechanism. These types of results with glass dissolution were presented for the first time during the course of writing this thesis. Based on the initial dissolution patterns obtained with the novel approach using TRIS, the experimental glasses could be divided into four distinct categories. The initial dissolution patterns of glasses correlated well with the anticipated bioactivity. Moreover, the normalized surface-specific mass loss rates and the different in vivo models and the actual in vivo data correlated well. The results suggest that this type of approach can be used for prescreening the suitability of novel glass compositions for future clinical applications. Furthermore, the results shed light on the possible bioactivity of glasses. An additional goal in this thesis was to gain insight into the phase changes occurring during various heat treatments of glasses with three selected compositions. Engineering-type T-T-T curves for glasses 1-98 and 13-93 were stablished. The information gained is essential in manufacturing amorphous porous implants or for drawing of continuous fibers of the glasses. Although both glasses can be hot worked to amorphous products at carefully controlled conditions, 1-98 showed one magnitude greater nucleation and crystal growth rate than 13-93. Thus, 13-93 is better suited than 1-98 for working processes which require long residence times at high temperatures. It was also shown that amorphous and partially crystalline porous implants can be sintered from bioactive glass S53P4. Surface crystallization of S53P4, forming Na2O∙CaO∙2SiO2, was observed to start at 650°C. The secondary crystals of Na2Ca4(PO4)2SiO4, reported for the first time in this thesis, were detected at higher temperatures, from 850°C to 1000°C. The crystal phases formed affected the dissolution behavior of the implants in simulated body fluid. This study opens up new possibilities for using S53P4 to manufacture various structures, while tailoring their bioactivity by controlling the proportions of the different phases. The results obtained in this thesis give valuable additional information and tools to the state of the art for designing glasses with respect to future clinical applications. With the knowledge gained we can identify different dissolution patters and use this information to improve the tuning of glass compositions. In addition, the novel online analysis approach provides an excellent opportunity to further enhance our knowledge of glass behavior in simulated body conditions.

Exchange rate regulation, the behavior of exchange rates, and macroeconomic stability in Brazil

In the last two decades an entirely new set of rules governing the foreign exchange transactions was established in Brazil, substituting for the framework inherited from the 1930s. Foreign exchange controls were dismantled and a floating exchange rate regime replaced different forms of peg. In this paper we argue that although successful by comparison to previous experiences, the current arrangement has important flaws that should be addressed. We discuss how it first led to high volatility and extremely high interest rates, which, when overcome, gave way to a long lasting appreciation of the real exchange rate with adverse consequences to industry.