Dow Jones Industrial Average Research Articles

Twitter mood predicts the stock market

Behavioral economics tells us that emotions can profoundly affect individual behavior and decision-making. Does this also apply to societies at large, i.e. can societies experience mood states that affect their collective decision making? By extension is the public mood correlated or even predictive of economic indicators? Here we investigate whether measurements of collective mood states derived from large-scale Twitter feeds are correlated to the value of the Dow Jones Industrial Average (DJIA) over time. We analyze the text content of daily Twitter feeds by two mood tracking tools, namely OpinionFinder that measures positive vs. negative mood and Google-Profile of Mood States (GPOMS) that measures mood in terms of 6 dimensions (Calm, Alert, Sure, Vital, Kind, and Happy). We cross-validate the resulting mood time series by comparing their ability to detect the public's response to the presidential election and Thanksgiving day in 2008. A Granger causality analysis and a Self-Organizing Fuzzy Neural Network are then used to investigate the hypothesis that public mood states, as measured by the OpinionFinder and GPOMS mood time series, are predictive of changes in DJIA closing values. Our results indicate that the accuracy of DJIA predictions can be significantly improved by the inclusion of specific public mood dimensions but not others. We find an accuracy of 86.7% in predicting the daily up and down changes in the closing values of the DJIA and a reduction of the Mean Average Percentage Error (MAPE) by more than 6%.

The asymmetric impact of firm-specific and of index returns on the volatility processes of individual stocks

This article examines the volatility processes of the 30 constituent stocks of the Dow Jones Industrial Average (DJIA) from 1998 to 2007. Estimating the standard Glosten, Jagannathan and Runkle (GJR) model across the DJIA's components confirms previous empirical findings of individual stocks’ conditional variances being less asymmetric than that of the parent index. A modified specification is then tested, termed the GJR-I, where lagged signed market returns have replaced firm-specific returns. The results suggest that individual stock volatility is significantly correlated with past signed index returns and that the asymmetry phenomenon is more pronounced with respect to market news compared to firm-specific news. This result still holds after estimating an extended specification where the conditional variance responds both to idiosyncratic and systematic innovations. The fact that individual stock volatility responds more asymmetrically to market returns than to firm specific returns stands in contrast ...

Influence of the Time Scale on the Construction of Financial Networks

BackgroundIn this paper we investigate the definition and formation of financial networks. Specifically, we study the influence of the time scale on their construction.Methodology/Principal FindingsFor our analysis we use correlation-based networks obtained from the daily closing prices of stock market data. More precisely, we use the stocks that currently comprise the Dow Jones Industrial Average (DJIA) and estimate financial networks where nodes correspond to stocks and edges correspond to none vanishing correlation coefficients. That means only if a correlation coefficient is statistically significant different from zero, we include an edge in the network. This construction procedure results in unweighted, undirected networks. By separating the time series of stock prices in non-overlapping intervals, we obtain one network per interval. The length of these intervals corresponds to the time scale of the data, whose influence on the construction of the networks will be studied in this paper.Conclusions/SignificanceNumerical analysis of four different measures in dependence on the time scale for the construction of networks allows us to gain insights about the intrinsic time scale of the stock market with respect to a meaningful graph-theoretical analysis.

A Historical Perspective on Market Risks Using the DJIA Index Over One Century

AbstractThe availability of the Dow Jones Industrial Average (DJIA) index over one century allows an investigation of the stability through time of the statistical properties of this time series, and a measurement of the performance of the main risk evaluation methodologies. The main risk methodologies are presented and discussed. The difference between fundamental limitations (that cannot be changed) and particular implementations (that can be tested and improved) is clearly made. Thanks to the large sample (28850 days), clear conclusions can be reached with simple statistics. In particular, the various risk methodologies perform just according to their respective ingredients used to build the volatility forecast (short versus long memory) and distributional assumption (normal versus fat‐tail). The time stability analysis is carried out using the algorithms developed for risk evaluations. After the volatility is properly discounted, the statistical properties of the DJIA index appear to be extraordinarily stable over time, despite the very significant changes that occurred during the 20th century.

Identifying common dynamic features in stock returns

This paper proposes volatility and spectral based methods for the cluster analysis of stock returns. Using the information about both the estimated parameters in the threshold GARCH (or TGARCH) equation and the periodogram of the squared returns, we compute a distance matrix for the stock returns. Clusters are formed by looking to the hierarchical structure tree (or dendrogram) and the computed principal coordinates. We employ these techniques to investigate the similarities and dissimilarities between the ‘blue-chip’ stocks used to compute the Dow Jones Industrial Average (DJIA) index.

Performance of DJIA Stocks Using Fundamental Indexation

During the decade beginning on December 31, 1999, an equal-weighted portfolio of the Dow Jones Industrial Average (DJIA) stocks outperformed the market capitalization and priceweighted portfolios, with ending values of $1,020, $731, and $777 on the $1,000 originally invested, respectively. The net income and free cash flows-weighted portfolios performed the best with identical ending values of $1,082. The portfolio with the lowest standard deviation measure of total risk was weighted on net income (14.26%) and the portfolio with the highest total risk was weighted on total assets (27.78%). All portfolios had their worst performance in 2008, while the best performance varied across portfolios but were limited to the years 2003, 2006 and 2009.

An Analysis of the Random Walk Hypothesis based on Stock Prices, Dividends, and Earnings

This paper explores the stationarity of price movements, dividend yields, and earnings yields for stock market indices and individual stocks within the broader context of the random walk hypothesis. In general, in order for a stock’s price to follow a random walk, its future price must be unforecastable based on all currently available information in the stock market, including its price history. If a stock price is stationary in a given time period, its statistical process does not change over time, meaning that the series has a deterministic trend, which could even be flat. This investigation tests for stationarity in the time series of prices and dividend yields of the Dow Jones Industrial Average (DJIA), the S&P 500 Index, and their underlying component stocks based on the results of univariate and panel unit root tests. I also test for the stationarity of earnings yields for the components of the DJIA. I find that prices of the DJIA and its underlying components behave in a more stationary manner than do the prices of the S&P 500 and its underlying components. Dividend yields behave in an equally non-stationary fashion for the underlying components of both the DJIA and S&P 500. Interestingly, earnings yields for the DJIA prove to exhibit more stationarity than the dividend yields for the DJIA and S&P 500, suggesting that earnings data have some predictability for stock prices.

Financial Markets in Motion: Visualising stock price and news interactions during the 2008 global financial crisis

Abstract This paper introduces a visualisation technique to help analyse large sets of financial data. The dataset used consists of price and news headlines for the Dow Jones Industrial Average (DJIA) and related companies for several weeks prior to the height of the Global Financial Crisis (GFC). Our visualisation can help analysts to uncover relationships between news headlines that can convey meaning about trader sentiment, and their influence on stock prices. This information is displayed visually in an interactive and animated Microsoft Excel dashboard consisting of three charts. Improvements in scalability and performance over previous techniques enables concurrent analysis of stock prices and news headlines from large datasets. A video screen capture is available to showcase our system [1] .

Evolving Least Squares Support Vector Machines for Stock Market Trend Mining

In this paper, an evolving least squares support vector machine (LSSVM) learning paradigm with a mixed kernel is proposed to explore stock market trends. In the proposed learning paradigm, a genetic algorithm (GA), one of the most popular evolutionary algorithms (EAs), is first used to select input features for LSSVM learning, i.e., evolution of input features. Then, another GA is used for parameters optimization of LSSVM, i.e., evolution of algorithmic parameters. Finally, the evolving LSSVM learning paradigm with best feature subset, optimal parameters, and a mixed kernel is used to predict stock market movement direction in terms of historical data series. For illustration and evaluation purposes, three important stock indices, S&P 500 Index, Dow Jones Industrial Average (DJIA) Index, and New York Stock Exchange (NYSE) Index, are used as testing targets. Experimental results obtained reveal that the proposed evolving LSSVM can produce some forecasting models that are easier to be interpreted by using a small number of predictive features and are more efficient than other parameter optimization methods. Furthermore, the produced forecasting model can significantly outperform other forecasting models listed in this paper in terms of the hit ratio. These findings imply that the proposed evolving LSSVM learning paradigm can be used as a promising approach to stock market tendency exploration.

Statistical properties, dynamic conditional correlation and scaling analysis: Evidence from Dow Jones and Nasdaq high-frequency data

This paper investigates statistical properties of high-frequency intraday stock returns across various frequencies. Both time series and panel data are utilized to explore the properties of probability distribution, dynamic conditional correlations, and scaling analysis in Dow Jones Industrial Average (DJIA) and Nasdaq intraday returns across 10-min, 30-min, 60-min, 120-min, and 390-min frequencies. The evidence shows that both returns and volatility (standard deviation) increase with the increasing scaling from 10-min to 390-min intervals. By fitting an AR(1)-GARCH(1,1) model to intraday data, we find that AR(1) coefficients are negative for DJIA returns and positive for Nasdaq, exhibiting a positive and negative feedback strategy in DJIA and Nasdaq, respectively. The evidence also shows that these coefficients are statistically significant for either including or excluding opening returns for the 10-min and 30-min frequencies. By examining the dynamic conditional correlation between the DJIA and the Nasdaq across different frequencies, a positive correlation ranging from 0.6 to 0.8 was found. In addition, the variance of the dynamic correlation coefficients is decreasing and appears to be stable for the 2001–2003 period. Finally, both returns on DJIA and Nasdaq satisfy the stable Lévy distributions, implying that both markets can converge to equilibrium by self-governing mechanism after shocks. Results of this work provide relevant implications for investors and policy makers.

Trading strategy design in financial investment through a turning points prediction scheme

Turning points prediction has long been a tough task in the field of time series analysis due to its strong nonlinearity, and thus has attracted many research efforts. In this study, the turning points prediction (TPP) framework is presented and further employed to develop a novel trading strategy designing approach to financial investment. The TPP framework is a machine learning-based solution incorporating chaotic dynamic analysis and neural network modeling. It works on the ground of a nonlinear mapping deduced in financial time series through chaotic analysis. An event characterization method is created in TTP framework to characterize trend patterns in ongoing financial time series. The main contributions of this paper are (1) it presents an ensemble learning based TPP framework, within which the nonlinear mapping is approximated by the ensemble artificial neural network (EANN) model with a new parameters learning algorithm; (2) a genetic algorithm (GA) based threshold optimization procedure is described with a newly defined performance measure, named TpMSE, which is used as a cost function; and (3) a trading strategy designing approach is proposed based on the TPP framework. The proposed approach was applied to the two real-world financial time series, i.e., an individual stock quote time series and the Dow Jones Industrial Average (DJIA) index time series. Experimental results show that the proposed approach can help investors make profitable decisions.

An investigation of the weekend effect during different market orientations

Studies of stock returns over short horizons indicated irregularities in returns, the weekend effect, and consequently the notion of market efficiency has been questioned. Despite extensive research on the weekend effect, little research has been conducted to define the prominence of the seasonal anomaly in Bear markets versus non-Bear markets. In the paper the weekend effect is investigated for daily returns in the Dow Jones Industrial Average (DJIA), the S&P 500, and the NASDAQ for Bear and non-Bear markets. Results support a weekend effect but only during non-Bear market orientations and a possible day-of-the-week effect during Bear and non-Bear markets.

Can idiosyncratic volatility help forecast stock market volatility?

This paper examines the predictive power of idiosyncratic volatility in the context of daily stock market volatility dynamics. Specifically, the relative performance of various models of market volatility is considered with respect to whether idiosyncratic volatility is excluded or included as an explanatory variable in such models. Using high frequency data covering the thirty stocks within the Dow Jones Industrial Average (DJIA) index, the results indicate that the inclusion of idiosyncratic volatility leads to significant in-sample and out-of-sample improvements in the fit of all the volatility models considered. These results are shown to be relatively robust to the loss function adopted by the forecaster, with reasonable forecast accuracy improvements available to such forecasters.

DJIA Index Membership and the Flow of Information Between Stock Returns

We contribute to the limited literature on interactions between individual stocks at the market microstructure level by utilizing an asynchronous multi-asset data integration methodology. For the periods before and after the Dow Jones Industrial Average (DJIA) revisions in 1999 and 2004, we investigate effects of membership in the DJIA on the changes in the relative speed of price adjustment to common information. Although, for the analyzed stocks, index membership in many cases is associated with a higher relative speed of price adjustment, it is neither a necessary nor a sufficient condition for having a lead in price movement.

Microscopic origin of non-Gaussian distributions of financial returns

In this paper we study the possible microscopic origin of heavy-tailed probability density distributions for the price variation of financial instruments. We extend the standard log-normal process to include another random component in the so-called stochastic volatility models. We study these models under an assumption, akin to the Born–Oppenheimer approximation, in which the volatility has already relaxed to its equilibrium distribution and acts as a background to the evolution of the price process. In this approximation, we show that all models of stochastic volatility should exhibit a scaling relation in the time lag of zero-drift modified log-returns. We verify that the Dow–Jones Industrial Average index indeed follows this scaling. We then focus on two popular stochastic volatility models, the Heston and Hull–White models. In particular, we show that in the Hull–White model the resulting probability distribution of log-returns in this approximation corresponds to the Tsallis (t-Student) distribution. The Tsallis parameters are given in terms of the microscopic stochastic volatility model. Finally, we show that the log-returns for 30 years Dow Jones index data is well fitted by a Tsallis distribution, obtaining the relevant parameters.

A common factor analysis for the US and the German stock markets during overlapping trading hours

We employ a bivariate common factor model to establish a permanent-transitory decomposition of two major stock indices (the Deutsche Aktienindex (DAX) for Germany and the Dow Jones Industrial Average (DJIA) for the United States). Using high-frequency data, we (1) identify a common trend shared by both indices, (2) find that the DJIA contributes up to 95 % to the total innovation of the common factor, (3) show that both markets adjust within minutes to a system-wide shock, and (4) verify by hypothesis testing that the DJIA is the driving force in the transatlantic system of stock indices.

Hausdorff clustering of financial time series

A clustering procedure is introduced based on the Hausdorff distance as a similarity measure between clusters of elements. The method is applied to the financial time series of the Dow Jones industrial average (DJIA) index to find companies that share a similar behavior. Comparisons are made with other linkage algorithms.

Statistical distribution and time correlation of stock returns runs

In this paper, we focus on the statistical features and time correlation of runs which is defined as a sequence of consecutive gain/loss (rise/fall) stock returns. By studying daily data of the Dow Jones industrial average (DJIA), we get the following points: firstly, the distribution of length and magnitude of stock returns runs both follow an exponential law; secondly, runs length do lack significant time correlation, while runs magnitude exhibit a slow decay of time correlation with long persistence up to several months, which implies existence of volatility clustering. We expect the above properties may add new members to the family of stylized facts about stock returns.

Intraday Price Discovery in the DJIA Index Markets

Abstract: This paper explores the dynamics of price discovery between the Dow Jones Industrial Average (DJIA) index and its three derivative products: the DIAMOND exchange‐traded fund (ETF), the floor‐traded regular futures, and the electronically traded mini futures. Even though the American Stock Exchange is the primary listing exchange for the ETF, the analysis indicates that the electronically traded ETF on the Archipelago (ArcaEx) electronic communications network dominates the price discovery process for DIAMOND shares. The E‐mini futures contribute the most to price discovery, followed by the ArcaEx DIAMOND. The DJIA index and regular futures contribute least to price discovery. The analysis is repeated using the derivatives of the S&P 500 index as a robustness check. The results indicate that multi‐market trading ensures greater pricing efficiency. Informed traders favor electronic trading because of immediate and anonymous trade execution.

Sample period selection and long-term dependence: New evidence from the Dow Jones index

This study employs the classical and modified rescaled adjusted range statistic (R/S statistic) to investigate the sensitivity of the long-term return anomaly observed on the Dow Jones Industrial Average (DJIA) to sample and method bias. Daily data from 1/1/1970 to 17/3/2004 is used with sub-periods identified based on sign shifts in the mean returns as well as the October 1987 crash. The return series are also filtered to accommodate autoregressive conditional heteroskedastic (ARCH) innovations and short-term dependencies. Hurst exponent and V-statistic values for each of the filtered series for the whole sample and sub-periods are estimated, while polynomial regression techniques are applied to plot the V-statistics. These plots show oscillating cycles of varying lengths. Overall, we find the null hypothesis of no long-term dependence is accepted for the whole sample and every sub-period using the modified rescaled range test, but not necessarily using the classical rescaled adjusted range test. The later test does, however, reveal episodes of both positive and negative dependence over the various sample periods, which have been reported by other researchers.