Search

This paper focuses on forecasting the price of Bitcoin, motivated by its market growth and the recent interest of market participants and academics. We deploy six machine learning algorithms (e.g., Artificial Neural Network, Support Vector Machine, Random Forest, k-Nearest Neighbours, AdaBoost, Ridge regression), without deciding a priori which one is the ‘best’ model. The main contribution is to use these data analytics techniques with great caution in the parameterization, instead of classical parametric modelings (AR), to disentangle the non-stationary behavior of the data. As soon as Bitcoin is also used for diversification in portfolios, we need to investigate its interactions with stocks, bonds, foreign exchange, and commodities. We identify that other cryptocurrencies convey enough information to explain the daily variation of Bitcoin’s spot and futures prices. Forecasting results point to the segmentation of Bitcoin concerning alternative assets. Finally, trading strategies are implemented.

This study combines a fundamental analysis of the rationale for conservative investors’ transactions, as well as long-term, low-risk strategies, and a technical analysis of the search for entry points into short-term, high-risk speculation. A hypothesis about the possible adaptation of high-risk foreign-exchange-market strategies to a low-risk stock market, based on a multi-timeframe analysis of the intersection of 3 EMA plus stochastic (a combination of three moving averages and a stochastic oscillator), is proven. The study’s modeling is based on walk-forward, blind simulation, cross procedure for realistically testing a hypothesis that can be performed in nine steps (Colby, 2003.) Colby’s algorithm Its subject is ordinary shares of Sberbank of Russia, and its analysis shows an absence of uncharacteristic movements in the chosen period of maximum volatility, from 2007 to the present. This analysis was conducted for two timeframes (more than five years for the trend direction and less than three years for the entry point). For the EMA, parameters were set at 20, 50, and 200; for stochastic parameters were set at 14, 3, and 3, 80/20. The “failure swing” reversal pattern and new support and resistance lines were detected. The study’s main conclusions are that the simultaneous use of three EMAs makes determining a corridor or a trend fairly reliable, as well as setting stop-losses. Moreover, the use of an oscillator is found not to always be reasonable; its main task is to confirm a signal. A stochastic oscillator with an explicit trend should not be analyzed for the whole period under consideration—only the last values should be considered. Moving averages and oscillators give fewer false signals on medium-term timeframes than on short-term timeframes. Due to a change in trend direction, identifying new (defined and correct) support and resistance lines is found to be necessary.

This paper investigates if and to what point it is possible to trade on news sentiment and if deep learning (DL), given the current hype on the topic, would be a good tool to do so. DL is built explicitly for dealing with significant amounts of data and performing complex tasks where automatic learning is a necessity. Thanks to its promise to detect complex patterns in a dataset, it may be appealing to those investors that are looking to improve their trading process. Moreover, DL and specifically LSTM seem a good pick from a linguistic perspective too, given its ability to “remember” previous words in a sentence. After having explained how DL models are built, we will use this tool for forecasting the market sentiment using news headlines. The prediction is based on the Dow Jones industrial average by analyzing 25 daily news headlines available between 2008 and 2016, which will then be extended up to 2020. The result will be the indicator used for developing an algorithmic trading strategy. The analysis will be performed on two specific cases that will be pursued over five time-steps and the testing will be developed in real-world scenarios.

Predicting stock-price remains an important subject of discussion among financial analysts and researchers. However, the advancement in technologies such as artificial intelligence and machine learning techniques has paved the way for better and accurate prediction of stock-price in recent years. Of late, Support Vector Machines (SVM) have earned popularity among Machine Learning (ML) algorithms used for predicting stock price. However, a high percentage of studies in algorithmic investments based on SVM overlooked the overfitting nature of SVM when the input dataset is of high-noise and high-dimension. Therefore, this study proposes a novel homogeneous ensemble classifier called GASVM based on support vector machine enhanced with Genetic Algorithm (GA) for feature-selection and SVM kernel parameter optimisation for predicting the stock market. The GA was introduced in this study to achieve a simultaneous optimal of the diverse design factors of the SVM. Experiments carried out with over eleven (11) years’ stock data from the Ghana Stock Exchange (GSE) yielded compelling results. The outcome shows that the proposed model (named GASVM) outperformed other classical ML algorithms (Decision Tree (DT), Random Forest (RF) and Neural Network (NN)) in predicting a 10-day-ahead stock price movement. The proposed (GASVM) showed a better prediction accuracy of 93.7% compared with 82.3% (RF), 75.3% (DT), and 80.1% (NN). It can, therefore, be deduced from the fallouts that the proposed (GASVM) technique puts-up a practical approach feature-selection and parameter optimisation of the different design features of the SVM and thus remove the need for the labour-intensive parameter optimisation.

The aims of this study are twofold. First, to determine the sign and magnitude of the skewness risk premium (SRP) in the Italian index option market using two procedures: (i) skewness swap contract...

The article is devoted to studying interaction between investor and trust manager. There is an information asymmetry between the investor and the trust manager, which makes it necessary to create a model that would allow the investor to overcome this asymmetry at least partially and get information about the quality of the service. In previous researches, the author proposed a theoretical model, describing this interaction with linear functions. In this paper, a model with nonlinear functions has been created and the conclusions of the linear and the nonlinear models have been compared. As a result, the conclusions of models with linear and nonlinear functions regarding the nature of the relationship between the commission and the factors used in the model were the same.

Forecasting stock prices is an extremely challenging job considering the high volatility and the number of variables that influence it (political, economical, social, etc.). Predicting the closing price provides useful information and helps the investor make the right decision. The use of deep learning and more precisely of recurrent neural networks (RNNs) in stock market forecasting is an increasingly common practice in the literature. Long Short Term Memory (LSTM) and Gated Recurrent Unit (GRU) architectures are among the most widely used types of RNNs, given their suitability for sequential data. In this paper, we propose a trading strategy designed for the Moroccan stock market, based on two deep learning models: LSTM and GRU to predict the closing price in the short and medium term respectively. Decision rules for buying and selling stocks are implemented based on the forecasting given by the two models, then over four 3-year periods, we simulate transactions using these decision rules with different settings for each stock. The returns obtained will be used to estimate an expected return. We only hold stocks that outperform a benchmark index (expected return > threshold). The random search is then used to choose one of the available parameters and the performance of the portfolio built from the selected stocks will be tested over a further period. The repetition of this process with a variation of portfolio size makes it possible to select the best possible combination of stock each with the optimized parameter for the decision rules. The proposed strategy produces very promising results and outperforms the performance of indices used as benchmarks in the local market. Indeed, the annualized return of our strategy proposed during the test period is 27.13%, while it is 0.43% for Moroccan all share Indice (MASI), 15.24% for the distributor sector indices, and 19.94% for the pharmaceutical industry indices. Noted that brokerage fees are estimated and subtracted for each transaction. which makes the performance found even more realistic.

Wind power producers participating in today's electricity markets face significant variability in revenue streams, with potential high losses mostly due to wind's limited predictability and the intermittent nature of the generated electricity. In order to further expand wind power generation despite such challenges, it is important to maximize its market value and move decisively towards economically sustainable and financially viable asset management. In this paper, we introduce a decision-making framework based on stochastic optimization that allows wind power producers to hedge their position in the market by trading physically settled options in futures markets in conjunction with their participation in the short-term electricity markets. The proposed framework relies on a series of two-stage stochastic optimization models that identify a combined trading strategy for wind power producers actively participating in both financial and day-ahead electricity markets. The proposed models take into consideration penalties from potential deviations between day-ahead market offers and real-time operation and incorporates different preferences of risk aversion, enabling a trade-off between the expected profit and its variability. Empirical analysis based on data from the Nordic region illustrates high efficiency of the stochastic model and reveals increased revenues for both risk neutral and risk averse wind producers opting for combined strategies.

The objective of this paper is to simulate the trading of the currency pair BTC/USD, investigating through the theory of the genetic algorithms the best sets of trading strategies, simulating through a realistic order book the bitcoin price formation, and reproducing a bitcoin price series that exhibits some stylized facts found in real-time price series. In this artificial market model two kinds of agents, Chartists and Random traders, perform trading. Chartists trade through the application of trading rules. Specifically, a part of Chartists trades applying the best sets of trading rules selected by a genetic algorithm that simulates a trading system, based on four technical analysis indicators, searching for parameters of each indicator that guarantee the highest profits in the training period; the remaining part trades applying trading rules choosing their parameters in a random way. On the contrary random trader's trade without applying any trading strategy, issuing in a random way sell or buy orders. Results show that the best sets of rules found to guarantee the highest profits both in the training and in the testing periods, and perform well also in the artificial market model where the Chartists who adopt the best sets of trading rules are able to achieve higher profits.

Biological market theory provides a conceptual framework to analyse trade strategies in symbiotic partnerships. A key prediction of biological market theory is that individuals can influence resource value – meaning the amount a partner is willing to pay for it – by mediating where and when it is traded. The arbuscular mycorrhizal symbiosis, characterised by roots and fungi trading phosphorus and carbon, shows many features of a biological market. However, it is unknown if or how fungi can control phosphorus value when exposed to abrupt changes in their trade environment. We mimicked an economic ‘crash’, manually severing part of the fungal network (Rhizophagus irregularis) to restrict resource access, and an economic ‘boom’ through phosphorus additions. We quantified trading strategies over a 3-wk period using a recently developed technique that allowed us to tag rock phosphate with fluorescing quantum dots of three different colours. We found that the fungus: compensated for resource loss in the ‘crash’ treatment by transferring phosphorus from alternative pools closer to the host root (Daucus carota); and stored the surplus nutrients in the ‘boom’ treatment until root demand increased. By mediating from where, when and how much phosphorus was transferred to the host, the fungus successfully controlled resource value.

We investigate the predictability of the G10 currencies with respect to lagged currency returns from the perspective of a U.S. investor using the maximally predictable portfolio (MPP) approach of Lo and MacKinlay (1997). We show that, out-of-sample, the MPP yields a higher Sharpe ratio, higher cumulative return and lower maximum drawdown than both a naïve equalweighted portfolio of the currencies and an equal-weighted portfolio of momentum trading strategies, and that a mean-variance investor would be willing to pay a performance fee to switch from the naïve and momentum portfolios to the MPP. The MPP has performed particularly well since the 2008 financial crisis, in contrast with the momentum portfolio, the value of which declined significantly over this period. Our results are robust to the estimation window length, the type and level of portfolio weight constraints and transaction costs.

The results of recent replication studies suggest that false positive findings are a big problem in empirical finance. We contribute to this debate by reviewing a sample of articles dealing with the short‐term directional forecasting of the prices of stocks, commodities, and currencies. Screening all relevant articles published in 2016 by one of the 96 journals covered by the Social Sciences Citation Index in the category “Business, Finance,” we select only those studies that use easily accessible data of daily or higher frequency. We examine each study in detail, from the selection of the dataset to the interpretation of the results. We also include empirical analyses to illustrate the shortcomings of certain approaches. There are three main findings from our review. First, the number of selected papers is very low, which is surprising even when the strict selection criteria are taken into account. Second, there are hardly any relevant studies that use high‐frequency data—despite the hype about financial big data and machine learning. Third, the economic significance of the findings—for example, their usefulness for trading purposes—is questionable. In general, apparently good forecasting performance does not translate into profitability once realistic transaction costs and the effect of data snooping are taken into account. Other typical problems include unsuitable benchmarks, short evaluation periods, and nonoperational trading strategies.

Hedge funds have traditionally served wealthy individuals and institutional investors with the promise of delivering protection of capital and uncorrelated positive returns irrespective of market direction, allowing them to better manage portfolio risk. However, the financial crisis of 2008 has heightened investor sensitivity to the high fees, illiquidity, lack of transparency, and lockup periods typically associated with hedge funds. Hedge fund replication products, or clones, seek to answer these challenges by providing daily liquidity, transparency, and immediate exposure to a desired hedge fund strategy. Nonetheless, although lowering cost and adding simplicity by using a common set of factors, traditional replication products might offer lower risk-reward performance compared to hedge funds. This research explores hedge fund replication further by examining the importance of constructing clones with specific factors relevant to each hedge fund strategy, and then compares the strategy specific clone risk and reward performance against both actual hedge fund performance and hedge fund clones constructed using a more general set of common factors. Testing shows that using strategy specific factors to replicate common hedge fund strategies can offer superior risk-reward performance compared to previous general model clones.

This article considers the problem of optimal liquidation of a position in a risky security quoted in a financial market, where price evolution are risky and trades have an impact on price as well ...

The notion of statistical arbitrage introduced in Bondarenko (2003) is generalized to statistical G-arbitrage corresponding to trading strategies which yield positive gains on average in a class of scenarios described by a σ-algebra G. This notion contains classical arbitrage as a special case. Admitting general static payoffs as generalized strategies, as done in Kassberger and Liebmann (2017) in the case of one pricing measure, leads to the notion of generalized statistical G-arbitrage. We show that even under standard no-arbitrage there may exist generalized gain strategies yielding positive gains on average under the specified scenarios. In the first part of the paper we prove that the characterization in Bondarenko (2003), no statistical arbitrage being equivalent to the existence of an equivalent local martingale measure with a path-independent density, is not correct in general. We establish that this equivalence holds true in complete markets and we derive a general sufficient condition for statistical G-arbitrages. As a main result we derive the equivalence of no statistical G-arbitrage to no generalized statistical G-arbitrage. In the second part of the paper we construct several classes of profitable generalized strategies with respect to various choices of the σ-algebra G. In particular, we consider several forms of embedded binomial strategies and follow-the-trend strategies as well as partition-type strategies. We study and compare their behavior on simulated data and also evaluate their performance on market data.

Forecasting of stock market returns is a challenging research activity that is now expanding with the availability of new data sources, markets, financial instruments, and algorithms. At its core, the predictability of prices still raises important questions. Here, we discuss the economic significance of the prediction accuracy. To develop this question, we collect the daily series prices of almost half of the publicly traded companies around the world over a period of ten years and formulate some trading strategies based on their prediction. Proper visualization of these data together with the use of the No Free Lunch theoretical framework gives some unexpected results that show how the a priori less accurate algorithms and inefficient strategies can offer better results than the a priori best alternatives in some particular subsets of data that have a clear interpretation in terms of economic sectors and regions. This work was supported in part by the Universidad Carlos III de Madrid under Strategic Action 2013/00199/002.

We investigate the relationship between insider trading and stock returns in firms with concentrated ownership. To this end, we employ data from East Asian countries which span the period January 2003 to May 2012. Consistent with the previous literature, we find a significantly negative relation between the selling activity of insiders and stock returns. However, contrary to studies which focus on highly developed markets, we find that the buying activity of insiders is also inversely related to future stock returns. Our analysis shows that top directors with higher ownership levels drive this result, suggesting that the trading activity of insiders is not always associated with profit-making motives and can be explained by their level of ownership. Furthermore, we demonstrate that a trading strategy which focuses solely on purchases made by top directors with high ownership levels yields negative returns. The paper has important implications for outside investors who mimic the trading activity of insiders with the aim to realise profits. Postprint

Although, according to uncovered interest rate parity, exchange rates should move so as to prevent the carry trade being systematically profitable, there is a vast empirical literature demonstrating the opposite. High interest currencies more often tend to appreciate rather than depreciate, as noted by Fama (1984). In this paper, we treat volatility as the critical state variable and show that positive returns to the carry trade are overwhelmingly generated in the low-volatility “normal” state, whereas the high-volatility state is associated with lower returns or with losses as currencies revert to the long run level approximated by their mean real exchange rate – in other words, purchasing-power parity (PPP) tends to reassert itself, at least to some extent, during periods of turbulence. We confirm these results by comparing the returns from three possible monthly trading strategies: the carry trade, a strategy which is long the undervalued and short the overvalued currencies (the “fundamental” strategy) and a mixed strategy which involves switching from carry trade to fundamentals whenever the previous period's volatility was in the top quartile. We are left with the anomalous (but apparently robust) result that the mixed strategy appears to generate higher returns and Sharpe ratios than either of the pure strategies.

The capitalization-weighted total relative variation $\sum_{i=1}^d \int_0^\cdot \mu_i (t) \mathrm{d} \langle \log \mu_i \rangle (t)$ in an equity market consisting of a fixed number $d$ of assets with capitalization weights $\mu_i (\cdot)$ is an observable and nondecreasing function of time. If this observable of the market is not just nondecreasing, but actually grows at a rate which is bounded away from zero, then strong arbitrage can be constructed relative to the market over sufficiently long time horizons. It has been an open issue for more than ten years, whether such strong outperformance of the market is possible also over arbitrary time horizons under the stated condition. We show that this is not possible in general, thus settling this long-open question. We also show that, under appropriate additional conditions, outperformance over any time horizon indeed becomes possible, and exhibit investment strategies that effect it.

Deep learning is an effective approach to solving image recognition problems. People draw intuitive conclusions from trading charts; this study uses the characteristics of deep learning to train computers in imitating this kind of intuition in the context of trading charts. The three steps involved are as follows: 1. Before training, we pre-process the input data from quantitative data to images. 2. We use a convolutional neural network (CNN), a type of deep learning, to train our trading model. 3. We evaluate the model's performance in terms of the accuracy of classification. A trading model is obtained with this approach to help devise trading strategies. The main application is designed to help clients automatically obtain personalized trading strategies., 30 pages, 41 figures

This paper examines how the size of the rolling window, and the frequency used in moving average (MA) trading strategies, affects financial performance when risk is measured. We use the MA rule for market timing, that is, for when to buy stocks and when to shift to the risk-free rate. The important issue regarding the predictability of returns is assessed. It is found that performance improves, on average, when the rolling window is expanded and the data frequency is low. However, when the size of the rolling window reaches three years, the frequency loses its significance and all frequencies considered produce similar financial performance. Therefore, the results support stock returns predictability in the long run. The procedure takes account of the issues of variable persistence as we use only returns in the analysis. Therefore, we use the performance of MA rules as an instrument for testing returns predictability in financial stock markets.

Optimal variance stopping (O.V.S.) problems are a new class of optimal stopping problems that differ from the classical ones because of their non linear (quadratic) dependence on the expectation op...

We examine whether investor mood, driven by World Health Organization (WHO) alerts and media news on dangerous infectious diseases, is priced in pharmaceutical companies' stocks in the United States. We argue that disease-related news (DRNs) should not trigger rational trading. We find that DRNs have a positive and significant sentiment effect among investors (on Wall Street). The effect is stronger (weaker) for small (large) companies, who are less (more) likely to engage in the development of new vaccines. A potential negative investor climate (on Main Street) – induced by disease-related fear – does not alter the positive sentiment effect., Highlights • We examine whether investor mood, driven by World Health Organization alerts on infectious diseases, is priced in pharmaceutical stocks in the U.S. • Disease-related fear has a negative and significant effect on returns on pharmaceutical stocks. • Disease-related news (DRN) has a positive and significant effect on returns on pharmaceutical stocks. • The effect of DRN persists over several days following a WHO alert on infectious diseases. • The effect of DRN is stronger for smaller companies and is weaker for larger companies. • Our findings give rise to profitable trading strategies.

One of the most important research streams in finance is to understand the determinants of stock market dynamics. Using a large amount of linguistic data regarding 960,808 posts during a 5‐year time period (from March 1, 2008, to August 31, 2013), we develop an economics blogs pessimism indicator (consid- ered as a proxy for either investor sentiment or risk aversion), and we show its validity to build performing trading strategies.

We show that when a continuous dark pool is added to a limit order book that opens illiquid, book and consolidated fill rates and volume increase, but spread widens, depth declines, and welfare deteriorates. The adverse effects on market quality and welfare are mitigated when book-liquidity builds but so are the positive effects on trading activity. All effects are stronger when traders’ valuations are less dispersed, access to the dark pool is greater, horizon is longer, and relative tick size larger.

This study examines whether the technical trading strategies can outperform the unconditional buy-and-hold strategy to forecast stock price movements and earn excess returns, after adjusting transaction costs, in emerging Colombo Stock Exchange (CSE). The study uses daily market closing prices of All Share Price Index (ASPI), which is a composite index to represent whole market, for twenty five years from January 1985 to December 2010. The variable length moving average method and fixed length moving average method with sixteen different rules is used as the methodology to analyze data. The empirical findings of the study confirmed that the moving average trading strategies have statistically significant predictive ability in explaining the market and capable of generating excess return to investors. However after considering transaction costs the excess return is negligible.DOI: http://dx.doi.org/10.4038/kjm.v2i2.6549 Kelaniya Journal of Management Vol.2(2) 2013:32-50

The paper is devoted to the study of the tools of technical analysis using computer technology. It analyzes the current conditions and the essence of the stock market, the basis of its functioning. The advantages of technical analysis are presented including accessibility and relative simplicity of its use. It analyzes the main indicators of the world stock market development and modern constant signs and tendencies of the process. It is revealed that the improvement of the technical analysis and increase of the forecast accuracy is a complex process. The prospects for the effective use of technical analysis in trading strategies for promoting analysis by using historical data of price changes for the asset under study by using computer software are determined. Back testing calculations are presented making clear the fact that the financial result would be given by a trading strategy based on the historical profitability of the asset.

In this paper we study whether the commodity futures market predicts the commodity spot market. Using historical daily data on four commodities—oil, gold, platinum, and silver—we find that they do. We then show how investors can use this information on the futures market to devise trading strategies and make profits. In particular, dynamic trading strategies based on a mean–variance investor framework produce somewhat different results compared with those based on technical trading rules. Dynamic trading strategies suggest that all commodities are profitable and profits are dependent on structural breaks. The most recent global financial crisis marked a period in which commodity profits were the weakest.

Research does not indicate a consensus on the relationship between idiosyncratic volatility and asset returns. Moreover, the role of cross sectional higher order moments in predicting market returns is relatively unexplored. We show that the cross sectional volatility measure suggested by Garcia et al. is highly correlated with alternative measures of idiosyncratic volatility constructed as variance of errors from the capital asset pricing model and the Fama French model. We find that cross sectional moments help in predicting aggregate market returns in some sample countries and also provide information for portfolio formation, which is more consistent for portfolios sorted on sensitivity to cross sectional skewness.

The authors investigate the biases in the backtested performance of “alternative beta” strategies using a unique sample of 215 trading strategies developed and promoted by global investment banks. Their results lend support to the cautions in the recent literature regarding backtest overfitting and lack of robustness in trading strategy performance during the “live” period (out of sample). The authors report a median 73% deterioration in Sharpe ratios between backtested and live performance periods for the strategies, and they establish a link between performance deterioration and strategy complexity, with the realized reduction in live versus backtested Sharpe ratios of the most complex strategies exceeding those of the simplest ones by over 30 percentage points. The robustness of strategy exposure to risk factors varies between asset classes and strategies; it appears reasonable in equity volatility and FX carry strategies but quite weak in the equity value strategy in particular.

This paper deals with stochastic modelling and short time prediction of share price development in Czech stock market. The aim of this research is to create such models which can be used for creating automatic trading strategies that will beat the market. Reliability of these models is being checked in three highly liquid shares from Prague Stock O2, CEZ and KB in seven years long period in years 2006–2012. We used Markov chain analysis for modelling. In our models a state space is defined on the basis of cumulative daily changes of share price and a state space with eight states is used. The state space is defined parametrically as a multiple of standard deviation of daily yields for each share. There were 14 parameters calculated in total and for each parameter nine trading strategies for all shares were applied. It means that 378 trading strategies were calculated. We succeeded in finding a set of compact state space models and in applying a compact group of trading strategies on these models which always beat the market when invested in portfolio. The average annual market yield was 3.6%. The average yield of our portfolios oscillates between 4.7% and 14.8%. Strategies overcame the market also even after including transaction costs. After including transaction costs in amount of 0.1% from the trade volume a decrease of average annual yields would occur in the range from 0.45 to 2.1 percentage points. We reached the best results in the sideway trend and in shares with less changing volatility. Conclusions of this research are in contradiction to the Efficient Market Hypothesis. Results indicate that Czech stock market is not effective in any of its form.

We consider the problem of measuring the performance of a dynamic strategy, rebalanced at a discrete set of dates, whose objective is that of replicating a claim in an incomplete market driven by a general multi-dimensional affine process. The main purpose of the paper is to propose a method to efficiently compute the expected value and variance of the hedging error of the strategy. Representing the pay-off the claim as an inverse Laplace transform, we are able to get semi-explicit formulas for strategies satisfying a certain property. The result is quite general and can be applied to a very rich class of models and strategies, including Delta hedging. We provide illustrations for the cases of interest rate models and Heston's stochastic volatility model.

This paper revisits the soybean crush spread arbitrage work of Simon (1999) by studying a longer time period, wider variety of entry and exit limits, and the risk-return relationship between entry and exit limits. The lengths of winning and losing trades are found to differ systematically, with winning trades significantly shorter on average than losing trades. Exiting trades near the 5- day moving average is shown to improve trade performance relative to a reversal of sign and magnitude from the entry spread. These results lead to trading rules designed to prevent lengthy trades; however, the profitability of trading rules is found to be unstable.

In this paper, the evolutionary algorithm for the optimization of Forex market trading strategies is proposed. The introduction to issues related to the financial markets and the evolutionary algorithms precedes the main part of the paper, in which the proposed trading system is presented. The system uses the evolutionary algorithm for optimization of a parameterized greedy strategy, which is then used as an investment strategy on the Forex market. In the proposed system, a model of the Forex market was developed, including all elements that are necessary for simulating realistic trading processes. The proposed evolutionary algorithm contains several novel mechanisms that were introduced to optimize the greedy strategy. The most important of the proposed techniques are the mechanisms for maintaining the population diversity, a mechanism for protecting the best individuals in the population, the mechanisms preventing the excessive growth of the population, the mechanisms of the initialization of the population after moving the time window and a mechanism of choosing the best strategies used for trading. The experiments, conducted with the use of real-world Forex market data, were aimed at testing the quality of the results obtained using the proposed algorithm and comparing them with the results obtained by the buy-and-hold strategy. By comparing our results with the results of the buy-and-hold strategy, we attempted to verify the validity of the efficient market hypothesis. The credibility of the hypothesis would have more general implications for many different areas of our lives, including future sustainable development policies.

In this paper we use generally applicable non-parametric methods in an attempt to sort out the possible sources of momentum in stock markets (behavioural theories or omitted risk factors). Specifically, we present the results of bootstrap analysis and stochastic dominance tests for the Spanish stock market. Our results from the bootstrap analysis are found to depend on the resampling method used (with or without replacement). Nevertheless, the various stochastic dominance techniques applied have led us to the same conclusion, namely, that the winner portfolio stochastically dominates the loser portfolio, which is not consistent with the general asset-pricing models developed for risk-averse investors. This suggests the interest of analysing theories that relax the unbounded rationality assumptions that support many of the classical asset pricing models.

Undiversifiable (or systematic risk) has long been an enemy of investors. Many countercyclical strategies have been developed to counter this. However, like all insurance types, these strategies are generally costly to implement, and over time can significantly reduce portfolio returns in long and extended bull markets. In this paper, we discuss an alternative technique, founded on the premise of physiological bias and risk-aversion. We take a behavioral discussion in order to contextualize the insurance like characteristics of option pricing and discuss how this can lead to a mispricing of the asymmetric relationship between the VIX and the S&amp, P 500. To test this, we perform studies in which we find statistical inefficiencies, thereby making it possible to implement a method of hedging index option premium in a way that has displayed no monthly drawdowns in bullish periods, while still providing large returns in major sell-offs. The three versions of the strategy discussed have negative betas to the S&amp, P 500, while exhibiting similar risk-adjusted excess returns over both bull and bear markets. Further, the performance generated over the entire period, for all three strategies, is highly statistically significant. The results challenge the weak form of the Efficient Market Hypothesis and provide evidence that the methods of hedging could be a valuable addition to an equity rich portfolio for the purpose of counteracting systematic risk.

Decision Support Systems, 90, ISSN:0167-9236, ISSN:1873-5797

In the mutual fund literature, it is an established fact that investors “chase past performance”. However, the opposite impact of flows on performance is widely discussed. Mainly, liquidity costs are held responsible for short-term erosion of performance, while high inflows enhance performance over longer horizons. I investigate this relation for various groups of equity, bond, and money market funds and find significant outperformance in high inflow funds over several months, especially for specific bond fund groups. In addition, I test whether this information can be exploited using simple investment strategies but find that the abnormal returns are too low to offset associated costs.

Trading generates not only information about the payoff of the assets traded, but also information about the traders themselves. Over time this information creates reputation. By using a unique dataset on the Treasury bond market, we derive a measure of reputation. This is then used to group dealers on the basis of their reputation and to analyze how they react to the reputation of other dealers. We show that the same type of trade, on the same asset, in the same market can generate different volume and volatility patterns depending on the type of dealers originating it. We also identify the “salient traders”. These traders, even if they do not originate the biggest volume of trade, have the highest impact on the market. These results have strong implications in terms of forecastability of future returns, volatility and overall trading volume because they show that most of the explanatory power of trades is due to salient traders.

Traditional vertically integrated power utilities around the world have evolved from monopoly structures to open markets that promote competition among suppliers and provide consumers with a choice of services. Market forces drive the price of electricity and reduce the net cost through increased competition. Electricity can be traded in both organized markets or using forward bilateral contracts. This article focuses on bilateral contracts and describes some important features of an agent-based system for bilateral trading in competitive markets. Special attention is devoted to the negotiation process, demand response in bilateral contracting, and risk management. The article also presents a case study on forward bilateral contracting: a retailer agent and a customer agent negotiate a 24h-rate tariff.

An algorithm for computing Dynamic Nash Equilibria (DNE) in an extended ver- sion of Kiyotaki and Wright (1989) (hereafter KW) is proposed. The algorithm com- putes the equilibrium pro?le of (pure) strategies and the evolution of the distribution of three types of assets across three types of individuals. It has two features that together make it applicable in a wide range of macroeco- nomic experiments: (i) it works for any feasible initial distribution of assets; (ii) it allows for multiple switches of trading strategies along the transitional dynamics. The algorithm is used to study the relationship between liquidity, production, and inequality in income and in welfare, in economies where assets fetch di¤erent returns and agents have heterogeneous skills and preferences. One experiment shows a case of reversal of fortune. An economy endowed with a low-return asset takes over a similar economy endowed with a high-return asset because, in the former economy, a group of agents abandon a rent-seeking trading behavior and increase their income by trading and producing more intensively. A second experiment shows that a reduction of market frictions leads both to higher income and lower inequality. Other experiments evaluate the propagation mechanism of shocks that hit the assets?returns. A key result is that trade and liquidity tend to squeeze income inequality. Keywords :Trading Startegies, Liquidity, Matching,Decentralized Markets JEL :C61, C63, E41, E27, D63

Reciprocal trade agreements, usually known under the generic name of countertrade (CT) have been traditionally seen as a form of bilateralism, and thus as an inefficient form of international exchange. Although contemporary trade theories do not fully explain the increasing prevalence of CT transactions, we will argue that it is possible to construct and use a third (hybrid) institutional form, which is congruent with the transaction-cost theories, and we will show how — under market imperfections — countertrade can reduce transaction costs while conserving the efficiency gains generated by these specific arrangements.

The electricity industry throughout the world, which has long been dominated by vertically integrated utilities, has experienced major changes. Deregulation, unbundling, wholesale and retail wheeling, and real-time pricing were abstract concepts a few years ago. Today market forces drive the price of electricity and reduce the net cost through increased competition. As power markets continue to evolve, there is a growing need for advanced modeling approaches. Accordingly, this article looks at using software agents to help manage the complexity of electricity markets, particularly retail markets. The article focuses on bilateral trading and describes some important features of an agent-based system for bilateral contracting. Special attention is devoted to the characteristics and negotiation behaviour of Buyer and Seller agents.

The novelty of the evolving electric power industry implies that researchers lack insight into numerous open problems. There is a growing need for advanced modeling approaches that simulate the behavior of electricity markets over time. Accordingly, this article looks at using software agents to help manage the complexity of electricity markets, particularly retail markets, towards ensuring long-term capacity sustainability. The article focuses on bilateral trading and describes some important features of an agent-based system for bilateral contracting with demand response. Special attention is devoted to two strategies for promoting demand response: a “volume management” strategy, for Buyer agents, and a “price management” strategy, for Seller agents.

This research proposes a new non-parametric approach to pairs trading based on renko and kagi constructions which originated from Japanese charting indicators and were introduced to academic studies by Pastukhov. The method exploits statistical information about the variability of the tradable process. The approach does not find a long-run mean of the process and trade towards it like other methods of pairs trading. The only assumption we need is that the statistical properties of the spread process volatility remain reasonably constant. The theoretical profitability of the method has been demonstrated for the Ornstein–Uhlenbeck process. Tests on the daily market data of American and Australian stock exchanges show statistically significant average excess returns ranging from 1.4 to 3.6% per month and annualized Sharpe ratio from 1.5 to 3.4. Refereed/Peer-reviewed

The paper presents a non-probabilistic approach to continuous-time trading where, in analogy to the binomial option-pricing model, terminal payoffs resulting from a given trading strategy are meaningful ‘state-by-state’, i.e., path-by-path. In particular, we obtain results of the form: “If a certain trading strategy is applied and if the realized price trajectory satisfies a certain analytical property, then the terminal payoff is.…” This way, derivation of the Black and Scholes formula and its extension become an exercise in the analysis of a certain class of real functions. While results of the above forms are of great interest if the analytical property in question is believed to be satisfied for almost all realized price trajectories (for example, if the price is believed to follow a certain stochastic process which has this property with probability 1), they are valid regardless of the stochastic process which presumably generates the possible price trajectories or the probability assigned to the set of all paths having this analytical property.

International audience; The present paper deals with a special Random Forest Data Mining technique, designed to alleviate the significant issue of high dimensionality in volatile and complex domains, such as stock market prediction. Since it has been widely acceptable that media affect the behavior of investors, information from both technical analysis as well as textual data from various on-line financial news resources are considered. Different experiments are carried out to evaluate different aspects of the problem, returning satisfactory results. The results show that the trading strategies guided by the proposed data mining approach generate higher profits than the buy-and-hold strategy, as well as those guided by the level-estimation based forecasts of standard linear regression models and other machine learning classifiers such as Support Vector Machines, ordinary Random Forests and Neural Networks.

Computational Economics, 41 (4), ISSN:0927-7099, ISSN:1572-9974

In this paper we compare the predictive ability of the Black-Scholes Formula (BSF) and Artificial Neural Networks (ANNs) to price call options by exploiting historical volatility measures. We use daily data for the S&P 500 European call options and the underlying asset and furthermore, we employ nonlinearly interpolated risk-free interest rate from the Federal Reserve board for the period 1998 to 2000.Using the best models in each sub-period tested, our preliminary results demonstrate that by using historical measures of volatility, ANNs outperform the BSF.In addition, the ANNs performance improves even more when a hybrid ANN model is utilized. Our results are significant and differ from previous literature. Finally, we are currently extending the research in order to: a) incorporate appropriate implied volatility per contract with the BSF and ANNs and b) investigate the applicability of the models using trading strategies.

In this paper we provide a price characterization of efficient consumption bundles in multiperiod economies with market frictions. Efficient consumption bundles are those that are chosen by at least one rational agent with monotonic state-independent and risk-averse preferences and a given future endowment. Frictions include dynamic market incompleteness, proportional transaction costs, short selling costs, borrowing costs, taxes, and others. We characterize the inefficiency cost of a trading strategy -the difference between the investment it requires and the largest amount required by any rational agent to obtain the same utility level - and we propose a measure of portfolio performance based on it. We also show that the arbitrage bounds on a contingent claim to consumption cannot be tightened based on efficiency arguments without restricting preferences or endowments. We examine the efficiency of common investment strategies in economies with borrowing costs due to asymmetric information, short selling costs, or bid-ask spreads. We find that market frictions generally change and typically shrink the set of efficient investment strategies, shifting investors away from well-diversified strategies into low cost ones, and for large frictions into no trading at all. Hence we observe strategies that become inefficient with market frictions, as well as strategies that are rationalized by market frictions.