Trading Strategy

Portfolio optimization is one of the central problems in finance, concerned with allocating capital across assets to achieve the best possible balance between risk and return. Since the development of Modern Portfolio Theory (MPT) by Markowitz in the 1950s, financial researchers and practitioners have sought methods to enhance diversification and improve the stability of returns. However, classical frameworks suffer from restrictive assumptions such as normally distributed returns, constant correlations, and market stationarity. These assumptions often fail in real-world conditions, leading to portfolios that are suboptimal in volatile and complex market environments. The growing availability of financial data and advancements in computational methods have opened the door for artificial intelligence (AI), particularly deep learning (DL) and deep reinforcement learning (DRL), to transform portfolio optimization. Unlike traditional econometric and machine learning models that primarily focus on return prediction, DL and DRL approaches offer the potential to directly optimize investor-relevant objectives such as the Sharpe ratio, Sortino ratio, or Conditional Value-at-Risk (CVaR). This shift in methodology aligns model training more closely with actual investment outcomes. Recent studies highlight this promise. Zhang, Zohren, and Roberts (2020) introduced a deep learning framework that directly maximizes the Sharpe ratio for exchange-traded fund (ETF) portfolios, bypassing the forecasting stage altogether and demonstrating superior performance during turbulent periods such as the COVID-19 pandemic. Similarly, Huang, Zhou, and Song (2024) developed a Sharpe-ratio-based Actor-Critic reinforcement learning framework capable of dynamically adjusting allocations in the CSI 300 Index, achieving stronger risk-adjusted returns than mean-variance optimization and risk parity benchmarks. Together, these works illustrate both the viability and advantages of risk-aware DL and DRL methods. This paper builds on these foundations by proposing a hybrid framework that combines multi-agent reinforcement learning with multi-objective optimization, targeting Sharpe, Sortino, and CVaR simultaneously. By integrating forecasting-driven features through recurrent and convolutional architectures with decision-making processes rooted in reinforcement learning, the proposed approach aims to deliver robust, adaptive, and risk-sensitive portfolio strategies. The study contributes to the growing literature on AI in finance by (1) expanding risk-adjusted objectives beyond single metrics, (2) leveraging multi-agent specialization for enhanced adaptability, and (3) systematically benchmarking performance against both classical and modern baselines.

This paper extends the standard feedback trading model of Sentana and Wadhwani (1992) by allowing the demand for shares by feedback traders to depend on sentiment. Our empirical analysis of three largest Exchange-Traded Fund (ETF) contracts in the U.S. suggests that there is a significant positive feedback trading in these markets and the intensity of which is generally linked to investor sentiment. Specifically, the level of feedback trading tends to increase when investors are optimistic. In addition, we find that the influence of sentiment on feedback trading varies across market regimes. These results are consistent with the view that feedback trading activity is largely caused by the presence of sentiment-driven noise trading. Overall, the findings are important in understanding the role of sentiment in investment behaviour and market dynamics and are of direct relevance to the regulators and investors in ETF markets.

portfolio decision system information supply; portfolio decision finding. We also want to note that the advantages of the proposed currency system of forecasting exchange rates and stock prices offer opportunities for an effective use of naIve strategies, named maximas here.

This study provides a comprehensive review of machine learning (ML) applications in the fields of business and finance. First, it introduces the most commonly used ML techniques and explores their diverse applications in marketing, stock analysis, demand forecasting, and energy marketing. In particular, this review critically analyzes over 100 articles and reveals a strong inclination toward deep learning techniques, such as deep neural, convolutional neural, and recurrent neural networks, which have garnered immense popularity in financial contexts owing to their remarkable performance. This review shows that ML techniques, particularly deep learning, demonstrate substantial potential for enhancing business decision-making processes and achieving more accurate and efficient predictions of financial outcomes. In particular, ML techniques exhibit promising research prospects in cryptocurrencies, financial crime detection, and marketing, underscoring the extensive opportunities in these areas. However, some limitations regarding ML applications in the business and finance domains remain, including issues related to linguistic information processes, interpretability, data quality, generalization, and the oversights related to social networks and causal relationships. Thus, addressing these challenges is a promising avenue for future research.

This paper investigates seasonal and cyclical patterns in Nigeria's equity market using the NGX All-Share Index (ASI) from 2012 to 2025. It used descriptive statistics to examine day-of-week, monthly, quarterly, and election-year patterns, alongside changes in volumes and downside risk captured through drawdowns. The results indicate that timing effects exist but are narrower than textbook anomalies: Wednesdays and late-year (Q4/December) periods consistently outperform, while Fridays and early Q1 (February-March) underperform. Contrary to conventional expectations, the "January effect" and "Santa Rally" are not robust in the NGX; instead, December and May emerge as the strongest months. Election cycles exert a powerful influence, with 2015 and 2019 posting steep annual losses (-12% to-17%), while 2023 broke the pattern with a +40% gain amid reforms and liquidity inflows. Drawdowns remain a structural feature, often reaching-40% to-55%, with an average recovery time of about three months (87 days). Volume patterns confirm liquidity clustering midweek and in select months, highlighting its role as a confirmation rather than predictive signal. This paper conclude that investors must pair seasonal insights with event awareness and disciplined drawdown management, as timing advantages alone are insufficient in Nigeria's high-volatility frontier market.

The continuous-time version of model of asset pricing with asymmetric information is studied, and generalized by allowing time-varying noise trading. From rather simple assumptions we are able to derive the optimal trade for an insider; the trading intensity satisfies a deterministic integral equation, given perfect inside information, which we give a closed form solution to. We use a new technique called forward integration in order to find the optimal trading strategy. This is an extension of the stochastic integral which takes account of the informational asymmetry inherent in this problem. The market makers' price response is found by the use of filtering theory. The novelty is our approach, which could be extended in scope.

The continuous-time version of Kyle's (Econometrica 53(6):1315-1336 model of asset pricing with asymmetric information is studied, and generalized in various directions, i.e., by allowing time-varying liquidity trading, and by having weaker a priori assumptions on the model. This extension is made possible by the use of filtering theory. We derive the optimal trade for an insider and the corresponding price of the risky asset; the insider's trading intensity satisfies a deterministic integral equation, given perfect inside information.

The continuous-time version of model of asset pricing with asymmetric information is studied, and generalized in various directions, i.e., by allowing time-varying noise trading, and by allowing the orders of the noise traders to be correlated with the insider's signal. From rather simple assumptions we are able to derive the optimal trade for an insider; the trading intensity satisfies a deterministic integral equation, given perfect inside information. We use a new technique called forward integration in order to find the optimal trading strategy. This is an extension of the stochastic integral which takes account of the informational asymmetry inherent in this problem. The market makers' price response is found by the use of filtering theory. The novelty is our approach, which could be extended in scope.

We introduce two complementary indices for tracking and mitigating relational subjectivity in human–AI dialogue: the Relational Subjectivity Index (RSI) and the Normalized Subjectivity Index (NSI*). RSI is a local, time-resolved signal that captures short-horizon predictability and stability of conversational responses, revealing when an agent drifts toward flattened, over-stable interaction patterns. NSI* is a calibrated composite that aggregates stability, predictability, and directionality under interpretable weights, providing a global view of how a system balances responsiveness with safe, consistent behavior across episodes. We show how both indices can be computed from logged interactions and calibrated against reference judgments 𝑟 (𝑡) to align the composite with product or research goals (e.g., nuance vs. safety).

Empirically, time-series analyses indicate that RSI co-moves with phases of high predictability and that NSI* improves agreement with reference signals after calibration. A “return-of-form” histogram complements the time-series by summarizing how often a system comes back to its typical stance versus exploring alternative responses. We outline practical uses: (i) telemetry for live monitoring and alerting, (ii) a calibration loop that periodically re-weights NSI* to reduce conversational flattening, (iii) policy shaping that conditionally adjusts decoding or prompt routing when RSI exceeds thresholds, and (iv) human-in-the-loop review for segments where NSI* and 𝑟 (𝑡) disagree.

Together, RSI and NSI* operationalize a measurable notion of relational subjectivity and offer concrete levers—monitoring, calibration, and policy interventions—to preserve diversity of stance without sacrificing reliability, helping prevent the long-run “flattening” that can degrade user experience and model capability.

A liquidity trader wishes to trade a fixed number of shares within a certain time horizon and to minimize the mean and variance of the costs of trading. Explicit formulas for the optimal trading strategies show that risk-averse liquidity traders reduce their order sizes over time and execute a higher fraction of their total trading volume in early periods when price volatility or liquidity increases. In the presence of transaction fees, traders want to trade less often when either price volatility or liquidity goes up or when the speed of price reversion declines. In the multi-asset case, price effects across assets have a substantial impact on trading behavior.

Directional Change is a new way to summarise and capture periodic price activities as well as significant changes in time series using an event-based time scale. Suggested by the scaling law, there exist high profit opportunities in the foreign exchange market thanks to the remarkably long length of currency coastline. This dissertation aims to provide new insights into the directional-change trading strategy in the foreign exchange market using high-frequency data with the implementation of algorithmic trading. We extend the trendfollowing directional-change trading strategy by incorporating ideas of the trailing stop and trend based on the concept of the scaling law, in an attempt to acquire more statistically uniformly positive results in terms of profitability as well as overall performance. We backtest on historical major currency pairs across multiple directional-change thresholds, and conduct a post-trade analysis to investigate the impact of the trailing stop as well as the effect of trend. Finally, we present a new discovery in the directional-change event strategy suggested by the results of the extended strategies.

Bitcoin is the first and most famous cryptocurrency. It is a virtual currency that is operated in a decentralized form using cryptographic strategies called blockchains. Although it has experienced significant market acceptance by traders and investors in recent years, it also suffers from volatility and riskiness. Technical analysis is one of the most powerful tools used for trading signals' allocation using some algorithmic strategies called technical indicators. In this research, a newly proposed multi-objectives decomposition-based particle swarm optimization algorithm is used to find the best parameter values for some technical indicators, which in turn generates the best trading signals for Bitcoin trading. In this context, three conflicting objectives have been used, i.e., the return on investment, the Sortino-ratio, and the number of trades. The proposed algorithm is compared to the original MOEA/D algorithm as well as the indicators using their original parameters. Results showed the superiority of the proposed algorithm during the training and testing periods over the other benchmarks.

There is a huge literature on the existence of risk premia in the foreign exchange market and its influence in explaining the divergence between the forward exchange rate and the subsequently realised spot exchange rate. In this paper, we seek to model directly the risk premium as a mean-reverting diffusion process. This is done by making use of the spot-forward price relationship and assuming a geometric Brownian process for the spot exchange rate. We are able to obtain a stochastic differential equation system for the spot exchange rate, the forward exchange rate and the risk premium which we estimate using Kalman filtering techniques. The model is then applied to the French Franc/USD and Japanese Yen/USD exchange rates from 1 January 1990 to 31 December 1998. For both currencies our main findings show (i) the persistence of substantial positive time variation in the forward risk premium and its alternating regimes; and (ii) the presence of a term structure of the forward risk premia.

This study aims to optimize algorithmic trading strategies using the relative strength index (RSI) and the moving average convergence divergence (MACD) indicators in the Vietnamese stock market. An automated trading system is constructed to optimize indicator parameters using multi-objective particle swarm optimization (PSO) over three objective functions: total return, win rate, and number of trades. The system employs simultaneous optimization of parameters and signal aggregation for developing the optimal selection strategy. Based on daily Vietnam index data from 2018 to 2024, the results show that the PSO method surpasses the differential evolution (DE) method in both returns and execution time. Additionally, the optimal selection strategy achieves superior performance compared to benchmark strategies. It also demonstrates the ability to adapt to the preferences of traders by selecting appropriate indicators. Traders can use the MACD indicator to seek higher profits, while the RSI indicator is more suitable for minimizing transaction costs in a volatile market.

The copyright of this thesis vests in the author. No quotation from it or information derived from it is to be published without full acknowledgement of the source. The thesis is to be used for private study or noncommercial research purposes only. Published by the University of Cape Town (UCT) in terms of the non-exclusive license granted to UCT by the author.

The copyright of this thesis vests in the author. No quotation from it or information derived from it is to be published without full acknowledgement of the source. The thesis is to be used for private study or noncommercial research purposes only. Published by the University of Cape Town (UCT) in terms of the non-exclusive license granted to UCT by the author.

A new approach to algorithmic trading system development is presented. This approach, Kernel Price Pattern Trading (KPPT P ), allows the practitioner to link the performance of a learned classifier (that predicts the occurrence of the price pattern P ) to the profitability of the system. A positive definite kernel based distance that tries to capture the drivers of the process of price patterns formation and some results about the profitability of the system are also presented.

In this study we forecast the term structure of FIBOR/EURIBOR swap rates by means of recursive vector autoregressive (VAR) models. In advance, a principal components analysis (PCA) is adopted to reduce the dimensionality of the term structure. To evaluate ex-ante forecasting performance for particular short, medium and long term rates and for the level, slope and curvature of the swap term structure, we rely on measures of both statistical and economic performance. Whereas the statistical performance is investigated by means of the Henrikkson-Merton statistic, the economic performance is assessed in terms of cash flows implied by alternative trading strategies. Arguing in favor of local homogeneity of term structure dynamics, we propose a data driven, adaptive model selection strategy to 'predict the best forecasting model' out of a set of 100 alternative implementations of the PCA/VAR model. This approach is shown to outperform forecasting schemes relying on global homogeneity of the term structure.

a a a a b c d a b c d Share Cite Previous Next Algorithmic trading has revolutionized financial markets, offering rapid and efficient trade execution. The integration of deep learning (DL) into these systems has further enhanced predictive capabilities, providing sophisticated models that capture complex, non-linear market patterns. This systematic literature review explores recent advancements in the application of DL algorithms to algorithmic trading with a focus on optimizing financial market predictions. We analyze and synthesize the key DL architectures, such as recurrent neural networks (RNN), long short-term memory (LSTM), convolutional neural networks (CNN), and hybrid models, to evaluate their performance in predicting stock prices, volatility, and market trends. The review highlights current challenges, such as data noise, overfitting, and interpretability, while discussing emerging solutions and future research directions. Our findings provide a comprehensive understanding of how DL reshapes algorithmic trading and its potential to improve decision-making processes in volatile financial environments.

We develop an analytically tractable model integrating the risk-shifting problem between bondholders and shareholders with the moral hazard problem between shareholders and the manager. The presence of managerial moral hazard exacerbates the risk-shifting problem. The exibility of the optimal contract allows shareholders to relax nancial constraints precisely when it is most valuable to do so, thus increasing shareholder appetite for risk-shifting. In fact, the model predicts a non-monotonic relation between risk-shifting and nancial distress, thereby reconciling seemingly contradictory empirical evidence. Moreover, rms with greater concern for moral hazard issue less debt and choose lower leverage. The model is qualitatively consistent with stylized facts on the survival of rms. Implications for business cycles are also considered.

Options have had to deal with an unfortunate history related to their speculative nature and the lack of regulation to eliminate fraud. However, their use could improve the welfare of all investors, through the promotion of complete markets. Furthermore, options convey information about anticipated future volatility, cash dividends and interest rates that help to improve the informational efficiency of the financial markets. Despite the advantages in the use of options, options markets only became important with the standarization of the contracts, which increased the liquidity of the markets and reduced transaction costs. This fact was clear in the U.S., where conventional markets were eclipsed by the introduction of organized, or registered, options markets.

PurposeTo investigate the performance of moving average trading rules in an emerging market context, namely that of the Jordanian stock market.Design/methodology/approachThe conditional returns on buy or sell signals from actual data are examined for a range of trading rules. These are compared with conditional returns from simulated series generated by a range of models (random walk with a drift, AR (1), and GARCH‐(M)) and the consistency of the general index series with these processes is examined. Sensitivity analysis of the impact of transaction costs is conducted and standard statistical testing is extended through the use of bootstrap techniques.FindingsThe empirical results show that technical trading rules can help to predict market movements, and that there is some evidence that (short) rules may be profitable after allowing for transactions costs, although there are some caveats on this.Originality/valueNew results for the Jordanian market; use of sensitivity analysis to i...

Risk-adjusted stock ranking criteria applicable when stock returns are not normally distributed are able to generate more profitable momentum strategies than those based on usual cumulative or total return criterion. These alternative risk-return ratio criteria conform to properties of coherent risk measures and, in different form, capture the risk of the tail distribution. Replacing the cumulative return by the risk-adjusted criterion, we also utilize the ratio as the objective function in the portfolio optimization problem and obtain optimal risky winner and loser portfolios. Our results are robust to transaction costs for both equal-weighted and optimized-weighted strategies. In particular, our alternative ratios outperform the cumulative return and the Sharpe ratio across all strategies measured by total realized return and independent performance measures over the observed period.

While an extensive body of literature has investigated the existence of the day-of-the-week anomaly in different stock markets globally, their findings can only provide implications for potential arbitrage opportunities for domestic investors in the investigated markets. We, therefore, add to these studies by investigating the possibility of international arbitrage activities using such an anomaly, after accounting for currency risk. Initially, we re-confirm the disappearance of the effect in the US market (S&P500), implying that US domestic investors can no longer exploit the day-of-the-week trading strategy in their home market. Further, we test whether investors who use the US dollar as the main trading currency (including US investors) can exploit the anomaly in foreign markets. We employ the daily values of representative indices and the national currencies of the three ASEAN countries (Singapore, Thailand and Malaysia) from 1995 to 2014. We find that the anomaly is evident in all three markets and can be exploited by foreign investors. Furthermore, the Thai exchange is the best investment destination for foreign investors with the highest returns and lowest currency risk. The profitability of this trading strategy is independent of economic activities and significantly dependent on the performance and conditions of the financial markets.

The main target of this paper is to discuss a short-term strategy for trading the monthly USA Nonfarm Employment Reports (NFP; Non-Farm Payrolls), by incorporating binary options and temporal warning dynamics & triggering trading functionalities (TTF). The proposed trading strategy is not a fully documented trading system, because it is derived, as well as it has been back-tested on USA Markets sample data (2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016) with an initial formal definition and documentation. The nonfarm employment reports are well known to create market volatility fostering short-term trading. Hence, a strategy based on binary options and these TTF functionalities offer great profit opportunities. The current paper contributes to corporate finance literature by examining, analyzing and defining these TTF functionalities. For this purpose, four categories of shareholders are regarded: The long-term investors, the short-term swing traders, the short-term momentary speculators, and the intraday speculators. Paper concludes that, in daily and intraday NFP trading, the short-term swing traders -if they incorporate binary options and apply the proposed TTF in their trading plans and strategies-are benefit at the expense of momentary and intraday speculators, while the long-term investors are not actually affected by the NFP release reports trading.

This paper proposes a transparent, hybrid deep learning pipeline for stock market prediction and trading signal generation, specifically applied to RVNL (Rail Vikas Nigam Limited) stock data from the Indian financial market. The model combines a Transformer-based LSTM with a rule-based decision layer and SHAP (SHapley Additive exPlanations) value analysis to enhance both predictive performance and interpretability. We integrate technical indicators such as MACD, EMA, RSI, and Bollinger Bands to capture price momentum and volatility patterns over time. To reduce false signals, a logical rule engine filters model outputs based on domain-specific thresholds. SHAP explainability ensures trust by visualizing the most influential features contributing to predictions. Experimental results show that our approach outperforms traditional strategies like Buy Hold, achieving a Sharpe ratio of 1.31 and maintaining maximum drawdown under 7.2%. We also present an ablation study confirming the contribution of each module to the overall performance. A candlestick-based prediction system is additionally implemented to provide real-time interpretability. Our model offers a robust framework that balances accuracy with transparency, making it suitable for retail investors and financial practitioners seeking actionable insights. This work contributes to the growing field of explainable AI in finance by aligning model intelligence with real-world decision-making.

There is a long tradition of research using computational intelligence, i.e. methods from artificial intelligence (AI) and machine learning (ML), to automatically discover, implement, and fine-tune strategies for autonomous adaptive automated trading in financial markets, with a sequence of research papers on this topic published at major AI conferences such as IJCAI and in prestigious journals such as Artificial Intelligence: we show evidence here that this strand of research has taken a number of methodological missteps and that actually some of the reportedly bestperforming public-domain AI/ML trading strategies can routinely be out-performed by extremely simple trading strategies that involve no AI or ML at all. The results that we highlight here could easily have been revealed at the time that the relevant key papers were published, more than a decade ago, but the accepted methodology at the time of those publications involved a somewhat minimal approach to experimental evaluation of trader-agents, making claims on the basis of a few thousand test-sessions of the trader-agent in a small number of market scenarios. In this paper we present results from exhaustive testing over wide ranges of parameter values, using parallel cloud-computing facilities, where we conduct millions of tests and thereby create much richer data from which firmer conclusions can be drawn. We show that the best public-domain AI/ML traders in the published literature can be routinely outperformed by a "sub-zerointelligence" trading strategy that at face value appears to be so simple as to be financially ruinous, but which interacts with the market in such a way that in practice it is more profitable than the well-known AI/ML strategies from the research literature. That such a simple strategy can outperform established AI/ML-based strategies is a sign that perhaps the AI/ML trading strategies were good answers to the wrong question.

The "ZIP" adaptive automated trading algorithm has been demonstrated to outperform human traders in experimental studies of continuous double auction (CDA) markets populated by mixtures of human and "software robot" traders. Previous papers have shown that values of the eight parameters governing behavior of ZIP traders can be automatically optimized using a genetic algorithm (GA), and that markets populated by GA-optimized traders perform better than those populated by ZIP traders with manually-set parameter values. This paper introduces a more sophisticated version of the ZIP algorithm, called "ZIP60", which requires the values of 60 parameters to be set correctly. ZIP60 is shown here to produce significantly better results in comparison to the original ZIP algorithm (called "ZIP8" hereafter) when a GA is used to search the 60-dimensional parameter space. It is also demonstrated here that this works best when the GA itself has control over the dimensionality of the search-space, allowing evolution to guide the expansion of the search-space up from 8 parameters to 60 via intermediate steps. Principal component analysis of the best evolved ZIP60 parameter-sets establishes that no ZIP8 solutions are embedded in the 60-dimensional space. Moreover, some of the results and analysis presented here cast doubt on previously-published ZIP8 results concerning the evolution of new 'hybrid' auction mechanisms that appeared to be improvements on the CDA: it now seems likely that those results were actually consequences of the relative lack of sophistication in the original ZIP8 algorithm, because "hybrid" mechanisms occur much less frequently when ZIP60s are used.

Aset kripto (cryptocurrency) pertama kali diperkenalkan pada tahun 2009 dengan hadirnya Bitcoin. Bitcoin diciptakan oleh sosok anonim bernama Satoshi Nakamoto. Tujuan utamanya adalah menciptakan bentuk uang digital yang tidak memerlukan perantara seperti bank atau pemerintah untuk mengelola transaksi. Aset ini beroperasi di atas teknologi yang disebut blockchain, yaitu sistem pencatatan digital yang tersebar di banyak komputer di seluruh dunia. Dengan sifatnya yang transparan, aman, dan tidak mudah dimanipulasi, aset kripto semakin diminati, terutama di tengah ketidakpercayaan terhadap sistem keuangan tradisional. Dalam satu dekade terakhir, ratusan bahkan ribuan aset kripto bermunculan, masing-masing membawa inovasi dan kegunaan yang berbeda. B. Perbedaan Antara Investasi dan Trading Kripto Banyak orang baru di dunia kripto bingung membedakan antara investasi dan trading. Investasi kripto berarti membeli aset digital seperti Bitcoin atau Ethereum dengan niat menyimpannya selama bertahun-tahun, menunggu nilainya naik secara signifikan. Biasanya, investor tidak terlalu memedulikan fluktuasi harga jangka pendek. Sebaliknya, trading kripto melibatkan pembelian dan penjualan aset dalam waktu yang lebih singkat-bisa hitungan jam, hari, atau minggu-untuk mengambil keuntungan dari perubahan harga. Trader aktif menganalisis grafik harga dan menggunakan berbagai strategi untuk memutuskan kapan waktu terbaik untuk masuk atau keluar dari pasar. C. Alasan Populernya Trading Kripto di Kalangan Pemula Ada beberapa alasan mengapa banyak pemula tertarik dengan trading kripto. Pertama, potensi keuntungannya sangat besar dalam waktu singkat, terutama karena harga kripto sangat fluktuatif. Kedua, proses mendaftar dan memulai trading sangat mudah, cukup dengan smartphone dan koneksi internet. Ketiga, modal awal yang dibutuhkan relatif kecil, bahkan ada yang bisa mulai dengan Rp50.000. Terakhir, pasar kripto buka 24 jam sehari tanpa hari libur, memberikan fleksibilitas waktu yang tinggi bagi siapa pun. II. Dasar-dasar Kripto A. Pengertian Aset Kripto (Cryptocurrency) Cryptocurrency atau aset kripto adalah mata uang digital yang dirancang untuk digunakan sebagai alat tukar. Berbeda dengan uang biasa, kripto tidak dicetak oleh bank sentral, melainkan dihasilkan melalui proses teknologi dan kriptografi. Semua transaksi dan kepemilikan dicatat di jaringan blockchain, yang membuatnya transparan dan aman. B. Teknologi Blockchain secara Singkat Blockchain dapat diibaratkan sebagai buku besar digital yang mencatat semua transaksi. Namun bedanya, buku besar ini tidak disimpan di satu tempat saja, melainkan disalin dan tersebar di

A Stock market collapse occurs when stock prices drop by more than 10% across all main indexes. Predicting a stock market crisis is difficult because of the increased volatility in the stock market. Stock price drops can be triggered by a variety of factors, including corporate results, geopolitical tensions, financial crises, and pandemic events. For scholars and investors, predicting a crisis is a difficult endeavor. We developed a model for the prediction of stock crisis using Hybridized Feature Selection (HFS) approach. Firstly, we went for the suggestion of the HFS method for the removal of stock’s unnecessary financial attributes. The Naïve Bayes approach, on the other hand, is used for the classification of strong fundamental stocks. In the third step, Stochastic Relative Strength Index (StochRSI) is employed to identify a stock price bubble. In the fourth step, we identified the stock market crisis point in stock prices through moving average statistics. The fifth is the pre...

This paper explores neural network-based approaches for algorithmic trading in cryptocurrency markets. Our approach combines multi-timeframe trend analysis with high-frequency direction prediction networks, achieving positive risk-adjusted returns through statistical modeling and systematic market exploitation. The system integrates diverse data sources including market data, on-chain metrics, and orderbook dynamics, translating these into unified buy/sell pressure signals. We demonstrate how machine learning models can effectively capture cross-timeframe relationships, enabling sub-second trading decisions with statistical confidence. The implementation addresses real-world challenges including data pipeline reliability, network latency, and operational risk management. Our results show consistent performance across various market conditions, with profit factor exceeding traditional approaches while maintaining manageable drawdown profiles.

Traders make trade decisions specifying entry, exit, and stop loss prices. Technicians often decide on entry, exit, and stop loss prices based on a predefined set of technical rules. In this paper, we employ a method based on grammatical evolution to coevolve technical rules for entry, exit, and stop loss for trading a London Stock Exchange (LSE) based stock in high frequency. We consider the case of two class of investors with risk averse, and loss preferences and build a partial trading frontier given the preferences considered. The performance of the rules evolved is compared to a publicly available trading system called the turtle trading system (TTS) and the best rules produced by our method outperforms TTS.

Models of stock price prediction have customarily utilized technical indicators alone to produce trading signals. In this paper, we construct trading techniques by applying machine-learning methods to technical analysis indicators and stock market returns data. The resulting prediction models can be utilized as an artificial trader used to trade on any given stock trade. Here the issue of stock trading decision prediction is enunciated as a classification problem with two class values representing the buy and sell signals. The stacking technique utilized in this paper is to assist trader with applying the proposed algorithms in their trading using random forest which was staked with different algorithms which incorporates Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM) and Neural Network (NN). The experimental results indicated that Top Layer of Random Forest (TRF) produced the best performance among all the algorithms compared. This is an indication that ...

With the high-paced change in the world, the market trends are changing fast and so do the technological innovations. The market is constantly moving and being affected by a number of external factors, making it difficult for investors to make decision. To solve this problem, a technique called Algorithmic Trading is being implemented widely. It refers to use of computer algorithms to make trading decisions without human intervention. In this paper, the algorithm used to predict the stock prices is the LSTM model which is a methodology of neural networks.The existing models are less accurate as they do not take technical factors into consideration. LSTM model, having error of only 0.00036%, overcomes this limitation and also takes temporal factor into consideration.

The success of mutual funds engaging in momentum and contrarian trading strategies is predicated on the identification of mispriced stocks. Stock investor sentiment betas capture salient characteristics that predispose stocks to mispricing. Funds engage in momentum and contrarian trading in equal proportions, but differ in the sentiment betas of the stocks in their portfolios. Momentum funds hold stocks with higher sentiment betas, and with a wider spread of betas compared to contrarian funds. Fund excess returns are strongly related to change in sentiment index, and the mean and spread of the sentiment betas of their stocks.

As distributed energy resource (DER) projects grow in popularity, there has been little focus on their potential to influence the dynamic stability of a transmission system. If implemented in large enough numbers, DER equipment may feasibly be leveraged to provide ancillary services such as spinning reserve, system inertia, or system stabilization. To determine the potential impacts of DER on future transmission grid stability, dynamic models of DER load combinations were created, reduced in order, and scattered throughout central and southem Califomia within an existing Western Systems Coordinating Council (WSCC) dynamic model. Evaluation of the composite WSCC-DER dynamic model allows correlations to be drawn between various DER properties and the transmission grid stability. Key find- ings show that increasing DER inertia tends to destabilize the transmis- sion system. The cause of this counterintuitive relationship requires further study but likely is related to the high impedance separating the DER from the transmission grid.

There is currently a paucity of literature focusing on the relationship between the actual actions of staff members, who perpetrate some form of computer abuse, and the organisational environment in which such actions take place. A greater understanding of such a relationship may complement existing security practices by possibly highlighting new areas for safeguard implementation. In addition, if insights are afforded into the actions of dishonest staff, prior to the actual perpetration of a crime, then organisations may be able to expand their preventive scope, rather than relying solely on technical safeguards to stop the actual commission of some form of computer abuse. To help facilitate a greater understanding of the offender/environment dynamic, this paper assesses the feasibility of applying criminological theory to the IS security context. More specifically, three theories are advanced, which focus on the offender's behaviour in a criminal setting. After opening with a description of the theories, the paper provides an account of the Barings Bank collapse. Events highlighted in the case study are used to assess whether concepts central to the theories are supported by the data. The paper concludes by summarising the major findings and discussing future research possibilities.

Several models exist for estimating volatility of stocks. In this paper, comparisons are made for the performance characteristics of seven volatility estimators using the data for eleven Banks from the Nigerian Stock Exchange (NSE) daily prices for the period 3 rd January 2006 to 31 st December 2008. The estimations computed are: Standard Deviation, Historical Close-to-Close, Parkinson, Garman-Klass, Rogers-Satchell, Modified Garman-Klass and Yang Zhang volatility estimators. The volatility computations for the estimators employed the open, high, low and close values of daily prices using 5, 10 and 20 days intervals with no overlapping. The Models are automated using Microsoft Visual Basic Express Edition with the volatilities output generated by the estimators further analysed using SPSS and Microsoft Excel software packages. The criteria used to evaluate the performances of these volatility estimators are the Mean Absolute Deviation (MAD), Standard Error (STDERR) and Efficiency. The Efficiency test compares the relative uncertainty of the various estimators using standard deviation as the benchmark while the MAD and STDERR are used to find the mean absolute deviation and the standard error of the estimators respectively. In terms of MAD and STDERR, the Parkinson model performs better than other estimators while the Garman-Klass performs better than other estimators in Efficiency. The only common finding is that the Standard Deviation estimator is the least performing of the estimators. Finally, the levels of correlation between volatility estimators are found to be very high.

Recent studies using data on social media and stock markets have mainly focused on predicting stock returns. Instead of predicting stock price movements, we examine the relation between Facebook data and investors' decision making in stock markets with a unique data on investors' transactions on Nokia. We find that the decisions to buy versus sell are associated with Facebook data especially for passive households and also for nonprofit organizations. At the same time, it seems that more sophisticated investors-financial and insurance institutions-are behaving independently from Facebook activities.

Recent studies using data on social media and stock markets have mainly focused on predicting stock returns. Instead of predicting stock price movements, we examine the relation between Facebook data and investors' decision making in stock markets with a unique data on investors' transactions on Nokia. We find that the decisions to buy versus sell are associated with Facebook data especially for passive households and also for nonprofit organizations. At the same time, it seems that more sophisticated investors-financial and insurance institutions-are behaving independently from Facebook activities.

Recent studies using data on social media and stock markets have mainly focused on predicting stock returns. Instead of predicting stock price movements, we examine the relation between Facebook data and investors' decision making in stock markets with a unique data on investors' transactions on Nokia. We find that the decisions to buy versus sell are associated with Facebook data especially for passive households and also for nonprofit organizations. At the same time, it seems that more sophisticated investors-financial and insurance institutions-are behaving independently from Facebook activities.

The views expressed in this Working Paper arc those of the author(s) and do not necessarily represent those of the IMP or IMP policy. Working Papers describe research in progress by the author(s) and are published to elicit comments and to further debate. This paper assesses the extent to which crashes in emerging market currencies are predictable using simple logit models based on lagged macroeconomic and financial data. To evaluate our model, we calculate trading strategies in which an investor goes long or short in the currency depending on whether crash probabilities are low or high. When we estimate the model on part of the data and then use the parameter estimates to generate predictions for the remainder of the sample, we find that substantial profits may be made. Furthermore, the model correctly forecasts major crashes even on an out-of-sample basis.

We investigate whether momentum or reversal is the dominant phenomenon in short horizon (one-to four-week) foreign exchange rate returns. We find, based on a broad sample of 63 emerging and developed market currencies, evidence of momentum rather than reversal. Momentum strategy returns are as large as 8% p.a. The short-term momentum effect appears to be robust. Returns are larger in the earlier sub-period but still exist in the more recent period. The strategies are also profitable when the USD is appreciating or depreciating but they perform better in business cycle expansions.

This paper by Swarajkumar Gawali, presents a comprehensive, logicdriven framework for quantitative investing. By integrating principles from the CFA curriculum with the technical rigor of the MITx MicroMasters in Data Science, it outlines a pipeline covering factor construction, predictive modeling, portfolio optimization, backtesting, and real-world deployment. The aim is to demonstrate how ethical, explainable, and statistically robust investment systems can be developed by retail or institutional practitioners.Also Special Thanks to Manipal Academy of Higher Education.

Monetary policy loosening and the associated impact on credit availability may have played a role in the present financial crisis. If such liquidity risk exists and is undiversifiable, then loose monetary policy should be associated with a risk premium. This ...

Stock market automated investing is an area of strong interest for the academia, casual, and professional investors. In addition to conventional market methods, various sophisticated techniques have been employed to deal with such a problem, such as ARCH/GARCH predictors, artificial neural networks, fuzzy logic, etc. A computational system that combines a conventional market method (technical analysis), genetic programming, and multiobjective optimization is proposed in this work. This system was tested in six historical time series of representative assets from Brazil stock exchange market (BOVESPA). The proposed method led to profits considerably higher than the variation of the assets in the period. The financial return was positive even in situations in which the share lost market value. Genetic programming • Multiobjective optimization • Technical analysis • Stock exchange market • Feature selection • BOVESPA B Frederico G. Guimarães

~ The paper analyses the employment effects of alternative trade strategies within the manufacturing sector. Estimates of effective protection rates show that Argentinian import substitution (1-S) policies have severely discriminated against exportable manufacturing industries. In turn these industries are shown to be at least 15% more labour-intensive than import-competing industries. Export promotion (EP) policies are characterized by low and relatively uniform protection to domestic and export sales together with the maintenance over time of a realistic exchange rate. The adoption of this strategy would imply a radical departure from I-S policies of escalated incentives and overvalued exchange rates which, as shown in the paper, have on average favoured capital-intensive industries.