Narrow Results

This study investigates the application of Machine Learning (ML) techniques to identify undervalued stocks, addressing the limitations of traditional investment strategies that rely heavily on fundamental analysis. As financial markets become more complex, characterized by volatility and information asymmetry, conventional valuation methods often struggle to capture these dynamics. In contrast, ML offers the ability to analyze large datasets and uncover intricate patterns, presenting a data-driven alternative for stock selection and portfolio optimization. A comprehensive predictive framework was developed, integrating traditional financial ratios with novel features derived from value investing principles and technical analysis. Several ML models — Random Forest, Long Short-Term Memory (LSTM), and Support Vector Machines — were assessed for their ability to predict high-return stocks. Performance metrics, including accuracy, precision, and recall, were used to evaluate model effectiveness. Among the models tested, the LSTM demonstrated the highest accuracy at 0.81, proving its robustness in identifying undervalued stocks. This research contributes to the growing body of literature on ML in finance by offering a practical framework that bridges theoretical concepts with real-world applications. The study also emphasizes the importance of refining ML algorithms to improve model interpretability and transparency, crucial for fostering trust in these systems. Future research should explore the use of ensemble methods and alternative data sources to further enhance prediction accuracy, while addressing challenges related to accountability in ML-driven investment strategies. This work advances the conversation around algorithmic trading and the future of data-driven finance.

We use a modified Cont–Bouchaud model to explore the effectiveness of trading breaks. The modifications include that the trading activity of the market participants depends positively on historical volatility and that the orders of the agents are conditioned on the observed mispricing. Trading breaks, also called circuit breakers, interrupt the trading process when prices are about to exceed a pre-specified limit. We find that trading breaks are a useful instrument to stabilize financial markets. In particular, trading breaks may reduce price volatility and deviations from fundamentals.

From the stock markets of six countries with high GDP, we study the stock indices, S&P 500 (NYSE, USA), SSE Composite (SSE, China), Nikkei (TSE, Japan), DAX (FSE, Germany), FTSE 100 (LSE, Britain) and NIFTY (NSE, India). The daily mean growth of the stock values is exponential. The daily price fluctuations about the mean growth are Gaussian, but with a nonzero asymptotic convergence. The growth of the monthly average of stock values is statistically self-similar to their daily growth. The monthly fluctuations of the price follow a Wiener process, with a decline of the volatility. The mean growth of the daily volume of trade is exponential. These observations are globally applicable and underline regularities across global stock markets.

The objective of this work is to analyze the Indice de Precios y Cotizaciones (IPC), which is the Mexican stock market index, by using several statistical tools in order to study the tendencies that can shed light on the evolution of the IPC towards a more efficient market. The methodology used is to apply the statistical tools to the Mexican index and compare the results with a mature and well-known market index such as the Dow Jones Industrial Average (DJIA). We employ an autocorrelation analysis, and the volatility of the indexes, applied to the daily returns of the closing price on a moving time window during the studied period (1980–2018). Additionally, we perform an order three permutation entropy analysis, which can quantify the disorder present in the time series. Our results show that there is evidence that the IPC has become more mature since its creation and that it can be considered an efficient market since around year 2000. The behavior of the several techniques used shows a similar behavior to the DJIA which is not observed before that year. There are some limitations mainly because there is no high frequency data that would permit a more detailed analysis, specifically in the periods before and after a crisis is located. Our conclusion is that since around the year 2000, the Mexican stock index displays the typical behavior of other mature markets and can be considered as one.

Multi-scale behaviors emerge in financial markets as complex systems. In this study, we intended to employ multi-scale Shannon entropy to trace the information transition of these phenomena, at different levels of Tehran stock market index (TEDPIX). The obtained results show that, in various magnitude scales and time scales, entropy Granger-causes TEDPIX index in terms of linear and nonlinear aspects. The results revealed that Granger causalities exist between entropy and TEDPIX. The causalities were linear in monthly (noise), quarterly (noise), semi-yearly (noise) and yearly (useful information) time spans; on the other hand, in quarterly (useful information) time span, the causalities were nonlinear. In this regard, one can conclude that entropy would be able to predict the market’s behavior.

Individual agents in financial markets decide based on market conditions, external news and information, and personal idiosyncrasies; from the collective action of such agents arise the unpredictable market dynamics. These actions affect the overall market efficiency, which measures how well the price reflects all available information. Here, we implement the self-organizing Ising model of Zhou and Sornette [Eur. Phys. J. B 55, 175 (2007)] to probe the efficiency of simulated financial markets under various conditions. Efficiency is parametrized by the dispersion of the generalized Hurst exponents obtained from multifractal detrended fluctuation analysis. Scanning different model parameter sets reveals the regimes of efficiency values in simulated markets that compare with those obtained from real-world data.

We study the various sectors of the Bombay Stock Exchange (BSE) for a period of eight years from January 2006–March 2014. Using the data of the daily returns of a period of eight years we investigate the financial cross-correlation co-efficients among the sectors of BSE and Price by Earning (PE) ratio of BSE Sensex. We show that the behavior of these quantities during normal periods and during crisis is very different. We show that the PE ratio shows a particular distinctive trend in the approach to a crash of the financial market and can therefore be used as an indicator of an impending catastrophe. We propose that a model of analysis of crashes in a financial market can be built using two parameters: (i) the PE ratio and (ii) the largest eigenvalue of the cross-correlation matrix.

The stability of portfolio investment in stock market crashes with Markowitz portfolio is investigated by the method of theoretical and empirical simulation. From numerical simulation of the mean escape time (MET), we conclude that: (i) The increasing number $(Np)$ of stocks in Markowitz portfolio induces a maximum in the curve of MET versus the initial position; (ii) A critical value of $Np$ in the behavior of MET versus the long-run variance or amplitude of volatility fluctuations maximumlly enhances the stability of portfolio investment. When $Np$ takes value below the critical value, the increasing $Np$ enhances the stability of portfolio investment, but restrains it when $Np$ takes value above the critical value. In addition, a good agreement of both the MET and probability density functions of returns is found between real data and theoretical results.

The roles of the trading time risks (TTRs) on stock investment return and risks are investigated in the condition of stock price crashes with Hushen300 data (CSI300) and Dow Jones Industrial Average (ˆDJI), respectively. In order to describe the TTR, we employ the escape time that the stock price drops from the maximum to minimum value in a data window length (DWL). After theoretical and empirical research on probability density function of return, the results in both ˆDJI and CSI300 indicate that: (i) As increasing DWL, the expectation of returns and its stability are weakened. (ii) An optimal TTR is related to a maximum return and minimum risk of stock investment in stock price crashes.

We investigate the stochastic resonance of periodic volatility in two financial markets with stock crashes for Dow Jones component stocks and Hang Seng index, based on the modified Heston model with an effective potential to describe the stock crashes. We introduce a cosine term to Heston model and develop a modified Heston model with periodic stochastic volatility for capturing the periodicity of the volatility process or volatility clustering which was observed in historical financial data sets. The proposed model was tested against Dow Jones industrial and Hang Seng index data. The experimental results demonstrate that the proposed model fits the historical data well when compared with the original Heston model. The signal power amplification (SPA) is calculated and studied to investigate the stochastic resonance of the proposed dynamic system. Experimental results suggest that: (i) optimal values of volatility parameters can be identified which maximize the effects of systematic and non-systematic randomness to the market periodicity; (ii) different values of correlation strength between noise sources will cause critical phenomenon and induce single or multiple resonances.

It is difficult to simulate the dynamical behavior of actual financial markets indexes effectively, especially when they have nonlinear characteristics. So it is significant to propose a mathematical model with these characteristics. In this paper, we investigate a generalized Weierstrass–Mandelbrot function (WMF) model with two nonlinear characteristics: fractal dimension D where 2 > D > 1.5 and Hurst exponent (H) where 1 > H > 0.5 firstly. And then we study the dynamical behavior of H for WMF as D and the spectrum of the time series γ change in three-dimensional space, respectively. Because WMF and the actual stock market indexes have two common features: fractal behavior using fractal dimension and long memory effect by Hurst exponent, we study the relationship between WMF and the actual stock market indexes. We choose a random value of γ and fixed value of D for WMF to simulate the S&P 500 indexes at different time ranges. As shown in the simulation results of three-dimensional space, we find that γ is important in WMF model and different γ may have the same effect for the nonlinearity of WMF. Then we calculate the skewness and kurtosis of actual Daily S&P 500 index in different time ranges which can be used to choose the value of γ. Based on these results, we choose appropriate γ, D and initial value into WMF to simulate Daily S&P 500 indexes. Using the fit line method in two-dimensional space for the simulated values, we find that the generalized WMF model is effective for simulating different actual stock market indexes in different time ranges. It may be useful for understanding the dynamical behavior of many different financial markets.

A connection between the notion of information and the concept of risk and return in portfolio theory is deduced. This succeeds in two steps: A general moment-return relation for arbitrary assets is derived, thereafter the total expected return is connected to the Kullback-Leibler information. With this result the optimization problem to maximize the expected return of a portfolio consisting of n subportfolios by moment variation under a given value-at-risk constraint is solved. This yields an ansatz to price information.

In the present work we have analyzed the financial Asian crisis of 1997, and its consequences on emerging markets. We have done so by means of a phase transition model originally presented by A. Johansen and D. Sornette [1].

We have analyzed the crashes on leading indices of Hong Kong (HSI), Turkey (XU100), Mexico (MMX), Brazil (Bovespa) and Argentina (Merval).

With the exception of Argentina's index, we were able to obtain optimum values for the critical date, corresponding to the most probable date of the crash.

We extend and adapt a class of estimators of the parameter H of the fractional Brownian motion in order to estimate the (time-dependent) memory function of a multifractional process. We provide: (a) the estimator's distribution when H ∈ (0,3/4); (b) the confidence interval under the null hypothesis H = 1/2; (c) a scaling law, independent on the value of H, discriminating between fractional and multifractional processes. Furthermore, assuming as a model for the price process the multifractional Brownian motion, empirical evidence is offered which is able to conciliate the inconsistent results achieved in estimating the intensity of dependence in financial time series.

We propose a class of Markovian agent based models for the time evolution of a share price in an interactive market. The models rely on a microscopic description of a market of buyers and sellers who change their opinion about the stock value in a stochastic way. The actual price is determined in realistic way by matching (clearing) offers until no further transactions can be performed. Some analytic results for simple special cases are presented. We also propose basic interaction mechanisms and show in simulations that these already reproduce certain particular features of prices in real stock markets.

We solve the problems of mean–variance hedging (MVH) and mean–variance portfolio selection (MVPS) under restricted information. We work in a setting where the underlying price process $S$ is a semimartingale, but not adapted to the filtration $𝔾$ which models the information available for constructing trading strategies. We choose as $𝔾={𝔽}^{\mathrm{\text{det}}}$ the zero-information filtration and assume that $S$ is a time-dependent affine transformation of a square-integrable martingale. This class of processes includes in particular arithmetic and exponential Lévy models with suitable integrability. We give explicit solutions to the MVH and MVPS problems in this setting, and we show for the Lévy case how they can be expressed in terms of the Lévy triplet. Explicit formulas are obtained for hedging European call options in the Bachelier and Black–Scholes models.

Understanding of factors like economic fundamentals or bubbles that normally determine the returns of stock in any emerging market such as the Thai stock market is essential for academic, investment planning and public policy reasons. An empirical study of the existence of rational speculative bubbles in the Thai stock market is undertaken by using the Weibull Hazard model. The conventional Weibull Hazard model is used as a benchmark model for other speculative bubble models. Empirical results suggest the presence of rational speculative bubbles in the Thai stock market, especially during the pre-crisis period. While rational speculative bubbles were not present immediately after the post-crisis period, some were observed a few years after the crisis. A possible explanation for such a result concerning rational speculative behaviour and bubbles in the emerging stock markets could be attributed to the presence of market imperfections in emerging stock markets, requiring institutional and policy developments to ensure efficient operation of the stock market.

In this paper we consider dynamic games with continuum of players which can constitute a framework to model large financial markets. They are called semi-decomposable games.

In semi-decomposable games the system changes in response to a (possibly distorted) aggregate of players' decisions and the payoff is a sum of discounted semi-instantaneous payoffs. The purpose of this paper is to present some simple properties and applications of these games. The main result is an equivalence between dynamic equilibria and families of static equilibria in the corresponding static perfect-foresight games, as well as between dynamic and static best response sets. The existence of a dynamic equilibrium is also proven. These results are counterintuitive since they differ from results that can be obtained in games with a finite number of players.

The theoretical results are illustrated with examples describing large financial markets: markets for futures and stock exchanges.

We focus on the stochastic description of the stock price dynamics. Thereby we concentrate on the Heston model and the Hull–White model. We derive the stationary probability density distribution of the variance of both models in the case of zero correlation coefficient. These distributions are used to calculate solutions for the logarithmic returns of the stock price for short time lags. Furthermore we apply the solutions of both models to the German tick-by-tick Dax data [1]. The data are from May 1996 to December 2001. We use the probability density distributions of the logarithmic returns, calculated out of the data, and fit these distributions to the theoretical distributions.

Recent empirical analyses have shown that liquidity fluctuations are important for understanding large price changes of financial assets. These liquidity fluctuations are quantified by gaps in the order book, corresponding to blocks of adjacent price levels containing no quotes. Here we study the statistical properties of the state of the limit order book for 16 stocks traded at the London Stock Exchange (LSE). We show that the time series of the first three gaps are characterized by fat tails in the probability distribution and are described by long memory processes.