Table 1 compares the predictability **of** the Markovian and non-Markovian approaches **of** the strategy for a range **of** transaction costs. The first column represents the transaction cost required **to** take a long or short position in the spread. The optimal barrier levels for the Markovian (4) and non-Markovian (3) master equations are given in columns 3 and 5. They are identical, meaning that the profit per trade, b2 — b1 — c, is the same for both strategies. However, the trade frequency, computed as a function **of** the first-passage **time** (1) **of** the log- price xt, is model-dependent. It is governed by the probability density function **of** the asset return λ(ξ) and also depends on the waiting-**time** distribution ψ (τ ) for the non-Markovian approach. The probability **of** breaching a barrier is higher during high activity than illiquid periods. The impact **of** the waiting-**time** distribution on the model precision is particularly important for liquid trading pairs offering a very low transaction cost. When the cost c = 0.1%, the Markovian model is a poor predictor **of** the real profit generated by the strategy. It forecasts an expected profit **of** 1696% for the period considered while the actual expected profit reaches 258%. For lower frequency trading the contribution **of** the waiting-**time** distribution ψ(τ ) is limited and the non-Markovian equation underestimates the real profit.

Mostrar mais
5 Ler mais

see [ 11 ] , is a stochastic process that, roughly speaking, describes the velocity **of** a massive Brownian particle under the influence **of** friction. It is stationary, Gaussian and Markovian, being the only nontrivial process that satisfies these three conditions, up **to** allowing linear transformations **of** the space and **time** variables. Over **time**, the process tends **to** drift towards its long-term mean: such a process is called mean-reverting. This process can be thought as a modification **of** the **random** **walk** in **continuous** **time**, or Wiener process, in which the properties **of** the process have been changed so that there is a tendency **of** the **walk** **to** move back towards a central location, with a greater attraction when the process is further away from the center. It can also be considered as the **continuous**-**time** analogue **of** the discrete-**time** AR (1) process.

Mostrar mais
7 Ler mais

Thus obtained residual data can be used for regression model- ing. However, a question arises if these data contain important information or they are simply a **random** noise. In the second case the modeling would be certainly unsuccessful. **To** answer this question some **statistical** tests can be applied. One **of** them is the non-parametric runs-test that verifies a randomness hypothesis for a binary-valued data sequence; in case **of** **time**-series, the signs **of** the residual values **of** the **time**-series. Another test as- sumes, that if the residual component is **of** **random** nature then it is characterized by a low autocorrelation coefficient. In particular, the autocorrelation **of** the first order is tested (i.e. for **time** lag equal **to** 1), using the Durbin-Watson test. The description **of** the both tests can be found in [7]).

Mostrar mais
5 Ler mais

A **continuous** survival analysis will be used **to** understand the residential customer churn in the FT industry (contractual settings) in Portugal. Let be a **continuous** non-negative **random** variable, which represents the survival **time** in days. Two key concepts in survival analysis are the survival and the hazard function. The survival function is the probability **of** an individual **to** survive beyond **time** and the hazard function is the instantaneous potential per unit **time** for the event occurrence (customer churn), given that the individual has survived up **to** **time** . Survival models can accom- modate both the proportional hazards (PH) and the accelerated failure **time** (AFT) forms. PH models assume that the hazard rates **of** any two individuals are proportional over **time** and that the hazard ratio is constant over **time**, and as such, the effect **of** any covariate in the hazard function is constant over **time**. AFT models assume that there is a constant non-negative acceleration factor that stretches out or shrinks survival times. AFT models are linear models **of** .

Mostrar mais
9 Ler mais

The water resources quality **continuous** monitoring is a complex activity. It generates extensive databases with **time** series **of** many variables and monitoring points that require the **application** **of** **statistical** methods for the information extraction. The **application** **of** **statistical** methods for frequency analysis **of** **time** series is linked **to** attending **of** the basic assumptions **of** randomness, homogeneity, independence, and stationarity. However, despite its importance, the veriication **of** these assumptions in water quality literature is unusual. Therefore, the present study tests the Upper Iguaçu basin water quality **time** series against the mentioned hypotheses. Rejection was observed in 15%, 26%, 51% e 31% for randomness, homogeneity, independence, and stationarity, respectively. The results evidenced the strong relation between monitoring strategy, data assessment and meeting **of** basic **statistical** assumptions for the analysis **of** water quality **time** series. Even with the existence **of** possible solutions for addressing those issues, the standard monitoring strategies, with irregular frequencies and lack **of** representativeness in relation **to** other periods, beyond commercial, act as an obstacle **to** their implementation.

Mostrar mais
13 Ler mais

So, may approach was **to** divide all the 260 stocks into 13 different groups (20 assets per group), using a simply percentile division **of** the different factors. Moreover we must take in mind that not all groups always had 20 assets. As it has been explained early, some assets and some factors for specific stocks in specific years cannot be retrieved from Bloomberg, which may bias the results. However, since these differences were minimal, I chose **to** neglect that fact and used the available information **to** construct the different groups, ignoring the assets that do not present data for a specific period **of** **time**.

Mostrar mais
25 Ler mais

We derive **statistical** **arbitrage** bounds for the buying and selling price **of** European derivatives under incomplete markets. In this paper, incompleteness is generated due **to** the fact that the market is dry, i.e., the underlying asset cannot be transacted at certain points in **time**. In particular, we re¯ne the notion **of** **statistical** **arbitrage** in order **to** ex- tend the procedure for the case where dryness is **random**, i.e., at each point in **time** the asset can be transacted with a given probability. We analytically characterize several properties **of** the **statistical** **arbitrage**- free interval, show that it is narrower than the super-replication in- terval and dominates somehow alternative intervals provided in the literature. Moreover, we show that, for su±ciently incomplete mar- kets, the **statistical** **arbitrage** interval contains the reservation price **of** the derivative.

Mostrar mais
57 Ler mais

[20] N.A. Kolmogorov, “Wienersche Spiralen und Einige Andere Interessante Kurven im Hilbertschen Raum,” Compte Rendus de l’Académie des Sciences de 1’Union des Républiques Soviétiques Socialistes, vol. 26, no. 2, pp. 115–118, 1940. [21] T.I. Laakso, V. Välimäki, M. Kaljalainen, and U.K. Laine, “Splitting the Unit Delay,” IEEE Signal Processing Magazine, vol. 13, no. 1, pp. 30–60, Jan. 1996. [22] P. Lanusse, J. Sabatier, R. Malti, P. Melchior, X. Moreau, and A. Oustaloup, “Past and Some Recent CRONE Group Applications **of** Fractional Differentia- tion,” Presentation at the Symposium on Applied Fractional Calculus (SAFC07), Industrial Engineering School (University **of** Extremadura), pp. 15–17, Oct. 2007, Badajoz, Spain.

Mostrar mais
8 Ler mais

We consider a stochastic pure exchange economy where a finite set I **of** agents live in a world **of** uncertainty from **time** 0 **to** **time** T . Uncertainty is modeled by a complete probability space (Ω, F, P). Each ω ∈ Ω is a state **of** nature which is a complete description **of** one possible realization **of** all exogenous sources **of** uncertainty from **time** 0 **to** **time** T . The sigma-field F is the col- lection **of** events which are distinguishable at **time** T and P is a probability measure on (Ω, F). The probability space (Ω, F, P) is endowed with a filtra- tion F = {F(t) : t ∈ [0, T ]} which represents the **time** evolution **of** the agents’ knowledge about the states **of** nature. We assume that F(0) is P-almost surely trivial and that F satisfies the usual conditions **of** right-continuity and complete- ness. A process is said optional if it is O-measurable where O is the sigma-field on Ω × [0, T ] generated by right-**continuous** F-adapted processes with left-limits.

Mostrar mais
24 Ler mais

This paper will aim **to** presents the applications **of** a **continuous**-**time** parameter estimation method for estimating structural parameters **of** a real bridge structure. For the purpose **of** illustrating this method two case studies **of** a bridge pile located in a highly seismic risk area are considered, for which the structural parameters for the mass, damping and stiffness are estimated. The estimation process is followed by the validation **of** the analytical results and comparison with them **to** the measurement data. Further benefits and applications for the **continuous**-**time** parameter estimation method in civil engineering are presented in the final part **of** this paper.

Mostrar mais
5 Ler mais

This paper reports the results **of** tests on the weak-form market efficiency applied **to** stock market indexes **of** France, Germany, UK, Greece, Portugal and Spain, from January 1993 **to** December 2007. We use a serial correlation test, a runs test, an augmented Dickey-Fuller test and the multiple variance ratio test proposed by Lo and MacKinlay (1988) for the hypothesis that the stock market index follows a **random** **walk**. The tests are performed using daily and monthly data for the whole period and for the period **of** the last five years, i.e., 2003 **to** 2007. Overall, we find convincing evidence that monthly prices and returns follow **random** walks in all six countries. Daily returns are not normally distributed, because they are negatively skewed and leptokurtic. France, Germany, UK and Spain meet most **of** the criteria for a **random** **walk** behavior with daily data, but that hypothesis is rejected for Greece and Portugal, due **to** serial positive correlation. However, the empirical tests show that these two countries have also been approaching a **random** **walk** behavior after 2003. (JEL G14; G15)

Mostrar mais
20 Ler mais

that is a special case **of** the two parameter Mittag-Leffler function that is a generalization **of** the exponential **to** what it reduces when ν = 1. This function is well studied {see [1], [7], [19], [51], [54]}. An interesting implementa- tion was done by Prof. Podlubny and can be found at the site **of** MatLab. It is an implementation **of** the two param- eter generalized Mittag-Leffler function with precision control—usage: mlf(alfa, beta, z, p). Equation (11) sug- gests us **to** work with the step response instead **of** the impulse response **to** avoid derivatives or working with non-regular functions near the origin.

Mostrar mais
8 Ler mais

A more general class **of** steganalysis techniques pioneered independently can be designed **to** work with any steganographic embedding algorithm, even an unknown algorithm. Such techniques have subsequently been called Universal Steganalysis techniques or Blind Steganalysis Techniques. Such techniques essentially design a classifier based on a training set **of** cover-objects and stego-objects arrived at from a variety **of** different algorithms. Classification is done based on some inherent”features” **of** typical natural images, which can get violated when an image undergoes some embedding process. Prediction accuracy can be interpreted as the ability **of** the measure **to** detect the presence **of** a hidden message with minimum error on average. The feature should be independent on the type and variety **of** images supplied **to** it[7].

Mostrar mais
10 Ler mais

The problem **of** Gambler's ruin, which reserves behave according **to** a simple **random** **walk**, is presented in a lot **of** textbooks about the stochastic processes theory in relation with Markov Chains, **Random** Walks, Martingales and even in other contexts. Very clear approaches **to** this subject are Billingsley [1] and Feller [2] that solve the problem through the classic first step analysis in order **to** obtain a difference equation. This proceeding is followed in this work. In alternative, Grimmett and Stirzaker [3] and Karlin and Taylor [4] present also resolutions through the Martingales Theory, as an applications’ example **of** the Martingales Stopping **Time** Theorem.

Mostrar mais
7 Ler mais

O The ONE (Opportunistic Network Environment) é um simulador de redes oportunísticas. O simulador está a ser desenvolvido pelos projetos SINDTN (Security Infrastructure for Delay Tolerant Network) (SINDTN 2011) e CATDTN (Connectivity, Applications, and Trials **of** Delay Tolerant Networking) e conta com o apoio da Nokia Research Center (Finland) (Nokia Research Center 2011). O simulador está a ser desenvolvido em Java e é um projeto open-source, estando o seu código disponível para download no site do projeto (http://www.netlab.tkk.fi/tutkimus/dtn/theone/, acedido a Outubro 2012). Ao nível da portabilidade, sendo o simulador uma aplicação em Java, o The ONE é bastante flexível uma vez que funciona sem problemas nos principais sistemas operativos da atualidade, Windows, Mac OSX e Linux. A última versão lançada do software foi a 1.4.1 e foi lançado em 31/01/2011.

Mostrar mais
152 Ler mais

Strand asymmetry is usually measured by AT and GC skews, as expressed by (A-T)/(A+T) and (G-C)/(G+C), respectively (Perna and Kocher, 1995). Positive AT skew values indicate more A than T and positive GC skew values more G than C, and vice versa. Strand asymmetries can be generally stable in some taxa such as vertebrates, in which the vast majority **of** genes show positive AT and negative GC skews. In vertebrates, only the ND6 gene shows the opposite compositional bias because its coding polarity is the opposite from all the other protein-coding genes. However, in invertebrates, strand specific asymmetries are more variable, mainly because gene rearrangements (particularly gene inversions) are much more frequent in this group. Likewise, in invertebrates, the H-strand origins **of** replication also switch their orientation in many taxa (Hassanin et al., 2005; Scouras and Smith, 2006). Since the strand bias is related with the orientation **of** the replication process, a reverse replication mechanism will also lead **to** an inverse compositional bias. For example, all available mitogenomes **of** crinoids (Echinodermata; Scouras and Smith, 2006) and some species in Cephalochordata (Kon et al., 2007) have generalized reversed strand asymmetry, when compared with the remaining mitogenomes **of** the same taxonomic group (phylum or subphylum, respectively). That occurs due **to** an inversion **of** the replication origin (the control region) and consequently **of** the replication direction (the H-strand became the L-strand and vice versa) (Hassanin et al., 2005; Kilpert and Podsiadlowski, 2006; Fonseca et al., 2008). In conclusion, compositional heterogeneities can also result from switches in coding polarities **of** genes and from reversal replication process **of** mitochondrial genomes. If a given gene (or group **of** genes) is inverted in some taxa, this will contribute **to** increase the heterogeneity **of** the patterns **of** sequence evolution within a dataset. In these scenarios, classical homogeneous models are not expected **to** capture the complex sequence evolution, resulting in misleading phylogenetic relationships. On the contrary, one expects that heterogeneous models can incorporate the **time**- and sequence-heterogeneity **of** the sequences and correctly infer the phylogenetic relationships between taxa.

Mostrar mais
136 Ler mais

dependence from the temperature **of** casting the bronze **to** the mould on Figure 6. From the introduced cross - section **of** casts from the probe TDAg, it results that it together with considerably grows up the depth **of** the contraction cavity with the growth **of** the temperature **of** casting, and what joins with this executed along its axis the volumetric contraction grows up, especially bronze B555 (Fig. 5a and 6). The bronze B10 is characterizes considerably smaller volumetric contraction (Fig. 5b and 6), however overheated 1180 °C above and cast **to** the hot plaster mould, in the conditions **of** the atmospheric pressure, it undergoes strong gassing with what considerable decrease **of** the depth **of** the contraction cavity joins (Fig. 5b 1200 ° C and Fig . 6). Zinc as high active metal in the relation **of** oxygen influences the lower- ing **of** the content **of** gases dissolved in the bronze B555. Consid- erably larger content Zn in the chemical composition **of** the bronze B555 (approx. 5%), in the comparison with the bronze B10 (**to** 0.5 %), it favours creation on the surface **of** the solidifica- tion bronze **of** the layer **of** oxides Zn and Cu, in the composition natural slags about the smaller mass density from the liquid bronze, making difficult chemical adsorption and dissolving the hydrogen and oxygen in the liquid bronze. Bronze B10 including first **of** all the admixture approx. 10% Sn, element **of** little active in the relation **to** oxygen, it absorbs from surroundings highly both the hydrogen as and the oxygen, what it brings in the conse- quence, together with the growth **of** the temperature **of** casting, **to** gassing the bronze.

Mostrar mais
6 Ler mais

This work in combination with industrial tests **of** casting welding show that the causes **of** high-temperature brittleness are the partial tears **of** the structure and the hot cracks **of** both the castings and the welded and padded joints. Such phenomena should be treated as irreversible failures caused by the process **of** crystallisation that is in the area **of** co-existence **of** the solid and liquid structural constituent. The assessment **of** the resistance **to** hot fractures was conducted on the basis **of** the transvarestriant trial. The transvarestriant trial consists in changing **of** strain during welding It was stated that the range **of** the high-temperature brittleness is very broad, which significantly limits the **application** **of** the welding techniques **to** join or mend the elements made **of** alloy ZRE-1. The brittleness is caused mainly by metallurgical factors, i.e., precipitation **of** inter-metal phases from the solid solution.

Mostrar mais
6 Ler mais

Considerando que a utilização de diferentes metodologias pode ser a razão da obtenção de diferentes conclusões entre os pesquisadores sobre a superioridade relativa do Lucro Líquido e do Fluxo de Caixa das Operações para prever os fluxos de caixa futuros, o objetivo desta pesquisa é avançar no estudo de Lustosa (2006), avaliando, mediante utilização de outra metodologia de previsão (**random** **walk** em vez de regressão linear), a eficácia da projeção ex ante de quatro medidas futuras de fluxo de caixa para um e dois anos à frente, a partir de medidas ex post do fluxo de caixa das operações, isolado e em conjunto com os accruals de curto prazo, e do lucro líquido contábil.

Mostrar mais
62 Ler mais

achine learning methods can be used **to** help design energy-efficient buildings reducing energy loads while maintaining the desired internal temperature. They work by estimating a response from a set **of** inputs such as building geometry, material properties, project costs, local weather conditions, as well as environmental impacts. These methods require a training phase which considers a dataset drawn from selected variables in the problem domain. This paper evaluates the performance **of** four machine learning methods **to** predict cooling and heating loads **of** residential buildings. The dataset consists **of** 768 samples with eight input variables and two output variables derived from building designs. The methods were selected based on exhaustive research with cross validation. Four **statistical** measures and one synthesis index were used for the performance assessment and comparison. The proposed framework resulted in accurate prediction models with optimized parameters that can potentially avoid modeling and testing various designs, helping **to** economize in the initial phase **of** the project.

Mostrar mais
13 Ler mais