Consecutive Trading Days Research Articles

ResNLS: An improved model for stock price forecasting

AbstractStock prices forecasting has always been a challenging task. Although many research projects adopt machine learning and deep learning algorithms to address the problem, few of them pay attention to the varying degrees of dependencies between stock prices. In this paper we introduce a hybrid model that improves stock price prediction by emphasizing the dependencies between adjacent stock prices. The proposed model, ResNLS, is mainly composed of two neural architectures, ResNet and LSTM. ResNet serves as a feature extractor to identify dependencies between stock prices across time windows, while LSTM analyses the initial time‐series data with the combination of dependencies which considered as residuals. In predicting the SSE Composite Index, our experiment reveals that when the closing price data for the previous five consecutive trading days is used as the input, the performance of the model (ResNLS‐5) is optimal compared to those with other inputs. Furthermore, ResNLS‐5 outperforms vanilla CNN, RNN, LSTM, and BiLSTM models in terms of prediction accuracy. It also demonstrates at least a 20% improvement over the current state‐of‐the‐art baselines. To verify whether ResNLS‐5 can help clients effectively avoid risks and earn profits in the stock market, we construct a quantitative trading framework for back testing. The experimental results show that the trading strategy based on predictions from ResNLS‐5 can successfully mitigate losses during declining stock prices and generate profits in the periods of rising stock prices.

Behavioral economics: who are the investors with the most sustainable stock happiness, and why? Low aspiration, external control, and country domicile may save your lives—monetary wisdom

Slight absolute changes in the Shanghai Stock Exchange Index (SHSE) corresponded to the city’s immediate increases in coronary heart disease deaths and stroke deaths. Significant fluctuations in the Shenzhen Stock Exchange Index (SZSE) corresponded to the country’s minor, delayed death rates. Investors deal with money, greed, stock volatility, and risky decision-making. Happy people live longer and better. We ask the following question: Who are the investors with the highest and most sustainable stock happiness, and why? Monetary wisdom asserts: Investors apply their deep-rooted values (avaricious love-of-money aspiration and locus of control, Level 2) as a lens to frame critical concerns in the proximal-immediate (Shanghai Stock Exchange Index changes, Level 1) and the omnibus-distal contexts (domicile: city vs. country, Level 2) to maximize expected utility (portfolio changes, Level 1) and ultimate serenity (stock happiness, Level 1). We collected multilevel data—the longitudinal SHSE and 227 private investors’ daily stock happiness and portfolio changes for 36 consecutive trading days in four regions of China. Investors had an average liquid asset of $76,747.41 and $54,660.85 in stocks. This study is not a “one-shot” game with “nothing at stake.” We classified Shanghai and Beijing as the city and Shenzhen and Chongqing as the country. Our cross-level 3-D visualization reveals that regardless of SHSE volatility, investors with low aspiration, external control, and country domicile enjoy the highest and most sustainable stock happiness with minimum fluctuations. Independently, investors with low aspiration, external control, and country domicile tend to make fewer portfolio changes than their counterparts. Behaviorally, less is more, debunking the myth—risky decisions excite stock happiness. Our longitudinal study expands prospect theory, incorporates attitude toward money, and makes robust contributions to behavioral economics and business ethics. We help investors and ordinary citizens make happy, healthy, and wealthy decisions. Most importantly, the life you save may be your own.

Fractional Stochastic Volatility Pricing of European Option Based on Self-Adaptive Differential Evolution

The option pricing model can predict the future trend of the financial market. In order to more accurately describe the changing process of the financial market, the Hurst index which can describe the characteristics of long-term memory is introduced into the traditional Heston model. Under the assumption that the underlying asset price follows fractional Brownian motion, the fractional stochastic volatility pricing of European option pricing model (Hurst-Heston model) is constructed, and the closed solution of the model is obtained according to the partial differential equation satisfied by the model. By analyzing the relationship between Hurst index and asset price, it is found that the movement process of asset price under the hypothesis of this model is more consistent with the real market change law, which verifies the rationality of the model. In the process of empirical analysis of SSE 50ETF put option data, Self-adaptive Differential Evolution algorithm is used to estimate parameters. The results showed that the error of Hurst-Heston model is smaller than other models, and the prediction error for consecutive trading days is similar. It showed that the pricing results of Hurst-Heston model are more accurate and stable.

Breaking VIX at open: Evidence of uncertainty creation and resolution

Abstract We decompose daily (close-to-close) changes of VIX into overnight (close-to-open) and trading-hour (open-to-close) changes. Consistent with the notion that non-trading creates uncertainty and trading resolves uncertainty, we find that on average VIX increases overnight and decreases during trading hours. More importantly, we document an important seasonality in VIX, i.e., the non-trading day effect. Overnight increase of VIX involving weekends or holidays is significantly higher than that over two consecutive trading days. We also document that VIX exhibits a clear pattern around pre-scheduled overnight and trading-hour macroeconomic announcement. Finally, we show that breaking VIX changes into overnight and trading-hour components and incorporating the non-trading day effect lead to not only significant improvements in in-sample fitting but also superior performance of out-of-sample-forecasting and active trading strategies.

The Effect of Preceding Sequences on Stock Returns

This study explores the effect of the gambler's fallacy on stock returns. I hypothesize that if during a number of consecutive trading days, a stock's return is positive (negative), then due to the gambler's fallacy, at least some of the investors may believe that the stock's price "has" to subsequently fall (rise), and thus, to increase their willingness to sell (buy) the stock, resulting in negative (positive) abnormal market-adjusted stock returns. Employing a large sample of daily stock price data, I was able to document that following relatively long sequences of positive (negative) stock returns, abnormal stock returns are on average significantly negative (positive), indicating the existence of the price pressure towards the return sign reversal. Moreover, the magnitude of the effect is stronger for longer return sequences. The effect is found to be more pronounced for smaller and more volatile stocks, and is robust to other relevant company - and stock-specific factors.

The Effect of Stock Return Sequences on Trading Volumes

The present study explores the effect of the gambler’s fallacy on stock trading volumes. I hypothesize that if a stock’s price rises (falls) during a number of consecutive trading days, then the gambler’s fallacy may cause at least some of the investors to expect that the stock’s price “has” to subsequently fall (rise), and thus, to increase their willingness to sell (buy) the stock, resulting in a stronger degree of disagreement between the investors and a higher-than-usual stock trading volume on the first day when the stock’s price indeed falls (rises). Employing a large sample of daily price and trading volume data, I document that following relatively long sequences of the same-sign stock returns, on the days when the sign is reversed, the trading activity in the respective stocks is abnormally high. Moreover, average abnormal trading volumes gradually and significantly increase with the length of the preceding return sequence. The effect is slightly more pronounced following the sequences of negative stock returns, and remains significant after controlling for other potentially influential factors, including contemporaneous and lagged actual and absolute stock returns, historical stock returns and volatilities, and company-specific events, such as earnings announcements and dividend payments.

The effect of non-trading days on volatility forecasts in equity markets

Weekends and holidays lead to gaps in daily financial data. Standard models ignore these irregularities. Because this issue is particularly important for persistent time series, we focus on volatility modelling, specifically modelling of realized volatility. We suggest a simple way of adjusting volatility models, which we illustrate on an AR(1) model and the HAR model of Corsi (2009). We investigate daily series of realized volatilities for 21 equity indices around the world, covering more than 15 years, and we find that our extension improves the volatility models—both in sample and out of sample. For HAR models and for consecutive trading days, the mean squared error decreased by 2.34% in average and for the QLIKE loss function by 1.41%.

Monetary Wisdom: How Do Investors Use Love of Money to Frame Stock Volatility and Enhance Stock Happiness?

Monetary intelligence asserts: individuals apply their money attitude to frame critical concerns in the context and strategically select certain options to achieve financial goals and ultimate happiness. Bridging the gap between stock volatility and behavioral economics, we collected longitudinal data from multiple sources and at multiple times: First, private investors (N = 229) in Shanghai—the financial capital of China—completed their love of money attitude measure (Rich-affect, Motivator-behavior, and Importance-cognition) and demographic variables in a survey. Second, we recorded daily Shanghai Stock Exchange Composite Index (“the Index”) for 30 consecutive trading days during the financial crisis in 2008—public records. Third, we text-messaged investors, collecting their daily Index Happiness, Stock Percentage (stocks/liquid assets), and Stock Happiness—private information. Here, investors illustrate: high Rich investors fret about low Index happiness, yet high Rich and high Importance investors boast high stock happiness, supporting the endowment effect and investor hubristic smirk. High Motivator investors quickly adjust their stock percentage/portfolio, suffering low Index happiness and low stock happiness. Gender moderates the relationship between the Index and Index happiness. Our panel data of intra-personal changes of stock happiness demonstrate investor monetary wisdom in the boom-and-bust cycles. Behaviorally, investor must become masters (but not slaves) of money and deactivate money as a Motivator. Curbing the desire to become Rich enhances happiness after gains (boom/risk aversion); appreciating money’s Importance bestows happiness after losses (bust/risk seeking). We expand prospect theory and offer implications to investor wealth, health, and happiness during financial crisis in particular as well as individual subjective well-being and happiness in general.

A Portfolio Strategy Using Glassdoor’s Business Outlook Ratings

I document that a simple portfolio strategy, selling stocks with worsening business outlook, provides significant abnormal returns. I construct a portfolio of firms one day after they experience a change in business outlook for all sample trading days. Over the sample of 56 consecutive trading days, 3,502 adjustments in business outlook took place. The daily average abnormal return, produced by shorting a worsening portfolio of firms, is 0.232 %. The accumulated raw return of this portfolio over the 56 day sample period is 4.3%. The purpose of this study is two-fold. I, not only show the opportunity for a profitable portfolio strategy, but I also look at the implications that significant portfolio returns have on the usefulness and quality of employee-sourced data.

Halo, horn, or dark horse biases: Corporate reputation and the earnings announcement puzzle

Abstract The primary goal of this study is to explore whether the halo, horn, or dark horse effects manifest in U.S. stock markets. To investigate the issue, an ideal empirical framework is newly created by combining a proxy for previous corporate perception, corporate reputation, with a well-known financial market anomaly, the earnings announcement puzzle. The interdisciplinary research presented in this study is original in its methodological design. Using companies listed and traded on the NASDAQ, NYSE, and AMEX between the first quarter of 2006 and the second quarter of 2011, the results show that investors' buying behaviors are influenced by significant horn and dark horse effects in the stock markets. Our findings suggest a short-term trading strategy that adopts long (short) positions in low-reputation stocks with the highest (lowest) class of earnings surprise over the three-day period surrounding the earnings announcements, generating a premium of 9.984%. On the basis of the price reversal effect, our results reveal a long-term trading strategy that adopts long (short) positions in low-reputation stocks with the lowest (highest) class of earnings surprises, starting from two days after the earnings announcement. The long-short positions for thirty, forty-five and sixty consecutive trading days will respectively generate premiums of 2.672%, 5.918%, and 7.602%.

Reference Price Updating, Adaptation, and Trading Volume

Trading an individual stock in mental accounting, crowd’s traders usually buy and sell in terms of narrowly defined gain and loss relative to a reference price about a behavioral value rather than a fundamental value on a trading day. A reference price is the price where trading volume or trading weight is the maximal across different prices on a trading day. The trading action generates a limited number of patterns about cumulative trading volume distribution over a price range because of mental accounting, loss aversion, trading weight, and coherence or agreement in trading. However, a reference price is updated to realize a significant gain and loss from time to time after a large imbalance on a trading day. We study how a past gain or loss elicits people to trade more or less frequently in a time series. We examine their adaptive behaviors to gain or loss in trading by analyzing correlation between realized gain or loss and change in total trading volume in any two consecutive trading days. Crowd’s traders update a reference price about the value of an individual stock via learning and generate a significant realized gain or loss. They adapt to the gain and loss realized by the updating via trading volume increase or decrease significantly in any two consecutive trading days. Our study makes distinction on two types of a reference price in multiple reference prices in trading: a reference price about a value in a time interval and a reference price about realized gain or loss in a time series. It also captures the coexistence of apparently contrary attitudes to risk from crowd’s traders in stock market: risk aversion from sellers and risk seeking from buyers in the gains of high probability; and risk seeking from sellers and risk aversion from buyers for the losses of high probability.

Analysis of Financial Time Sequence using Hidden Markov Models

Financial time sequence analysis has been a popular research topic in the field of finance, data science and machine learning. It is a highly challenging due to the extreme complexity within the sequences. Mostly existing models are failed to capture its intrinsic information, factor and tendency. To improve the previous approaches, in this paper, we propose a Hidden Markov Model (HMMs) based approach to analyze the financial time sequence. The fluctuation of financial time sequence was predicted through introducing a dual-state HMMs. Dual-state HMMs models the sequence and produces the features which will be delivered to SVMs for prediction. Note that we cast a financial time sequence prediction problem to a classification problem. To evaluate the proposed approach, we use Shanghai Composite Index as the dataset for empirically experiments. The dataset was collected from 550 consecutive trading days, and is randomly split to the training set and test set. The extensively experimental results show that: when analyzing financial time sequence, the mean-square error calculated with HMMs was obviously smaller error than the compared GARCH approach. Therefore, when using HMM to predict the fluctuation of financial time sequence, it achieves higher accuracy and exhibits several attractive advantageous over GARCH approach.

Back to the Future: An Approximate Solution for N Out of M Soft-Call Option

In convertible bond market, it is very common to protect the conversion privilege from being called away too soon by using soft-call constraint, or to protect the bond being converted too soon by using provision convert constraint. The first option will protect the bond holder; the second will be benefit to bond issuer. Both constrains have the common feature that the option can be exercise only when the underlying stock closes above a pre-set barrier for any n or more days over m consecutive trading days up to the exercise day. This feature brings challenge for pricing. This paper will propose an approximation solution by Looking Backward (LB) method. In order to illustrate the idea more clearly, I will focus on the Black model stock dynamic using binomial tree based on Cox-Ross-Rubinstein scheme. The results are compared with the exactly solution given by the author in [1]. The extension to other numerical method such as PDE with more general stock dynamic will also be discussed, and the numerical scheme will be laid out. The idea of the method can be applied to the pricing of other path dependent instruments in general.

Persistent Profit and Portfolio Adjustment

This study investigates investors’ behaviors on the ir portfolio adjustment after gain and loss. Using a unique data set of transaction data, the study supports the prospect theory that the investors will be more cautious after two consecuti ve gains. They reduce their portfolio’s systematic risk by giving less weight on large capi talization stocks. Investors fix their gain or loss relative to the purchasing price of the stocks . A number of literature show that investors’ risk pr eferences towards risk are not symmetric as being assumed in a traditional framework. Kaheneman and Tversky (1979) first proposed the prospect theory. The theory replaces utility function with value function that is concave on gain and convex on loss relative to the initial wealth to explain investors’ preference. The new function can explain both risk-taking and r isk-averse behaviors normally found in the market. Recent empirical studies that use experime ntal data and market data support the idea. Using a unique data set of Thai investors’ portfoli o, this study contributes in two areas. It is the first study that investigates gain or loss from two -consecutive trading. Secondly, it investigates the portfolio adjustment of investors after gain or loss. The study measures the adjustment in terms of systematic risk, diversification, and sent iment. The results confirm asymmetric impact on trading st rategy made by investors after gain or loss. The two-consecutive trading day framework allows us to confirm that the investors lower their portfolio’s risk after two consecutive profitable trading days. They seem to fix their reference points in judging gain or loss with the p urchasing price. Since one third of the portfolio is invested in large capitalization stock s, the result is in line with Kliger and Kudryavtsev (2008) who suggested that the reference point adaptation is mostly found in low capitalization stock with limited information. If the investors realize their gain in the first perio d and still have profitable trade in the second perio d, they will readjust weights of existing stocks in their portfolio to be more diversified rather th an buying a new stock. The impact is profound when their portfolio consists mainly of the large c apitalization stocks. The investors become

Approximating the Embedded M Out of N Day Soft-Call Option of a Convertible Bond: An Auxiliary Reversed Binomial Tree Method

A convertible bond (CB) with a right of m out of n day provisional call or soft-call becomes callable given that the underlying stock closes above a pre-set trigger price for any m or more days over the n consecutive trading days up to the current day. It is computationally challenge to value the contribution of this embedded option to the price of CB. This paper proposes an approximation based on the idea of an auxiliary reversed binomial (ARB) tree, and shows that the approach can be efficiently implemented under the Cox-Ross-Rubinstein parameterization. Two important insights emerge from ARB. First, the convertible bond is unconditionally callable at higher stock prices, but uniformly not callable at lower prices. Second, the effect of the soft-call is rather localized around the trigger price. Surprisingly, the simple One-Touch, or one out of one, approximation is found to yield CB prices that are very close to those from ARB, even though ARB is found to a better approximation in almost every aspect. Further numerical results suggest that in order to generate well-behaved CB prices, cautions need to be taken while designing the terms of soft-call. Being independent of the finite difference pricing grid, the proposed ARB tree can also be used in association with the tree method or Monte Carlo simulation, and could in principle be applicable to exotic derivatives with similar embedded options.

Closing the Question on the Continuation of Turn-of-the-Month Effects: Evidence from the S&P 500 Index Futures Contract

Prior research documents unusually high returns on the last trading day of the month and over the next three consecutive trading days. This phenomenon is known as the turn-of-the-month (TOTM) effect. According to Siegel (1998), why these anomalies occur is not well understood, and whether they will continue to be significant in the future is an open question. In this paper, we examine the S&P 500 futures contract for evidence that turn-of-the-month effects have continued. Transaction costs are low for index futures, and the absence of short-sale restrictions makes index futures an attractive venue for testing the continuation of market anomalies because of the low cost of arbitrage. We find that TOTM effects for S&P 500 futures disappear after 1990, and this result carries over to the S&P 500 spot market. We conjecture that a change in the preference of individual investors over time from making direct to making indirect stock purchases through mutual funds is related to the disappearance of the TOTM effect for more recent return data. In this paper, we argue that turn-of-the-month return patterns for both spot and futures prices are dynamic and related to market microstructure and therefore subject to change without notice. Financial economists should be careful when making out-of-sample inferences from observed in-sample return regularities.

Genetic Programming Prediction of Stock Prices

Based on predictions of stock-prices using genetic programming (or GP), a possibly profitable trading strategy is proposed. A metric quantifying the probability that a specific time series is GP-predictable is presented first. It is used to show that stock prices are predictable. GP then evolves regression models that produce reasonable one-day-ahead forecasts only. This limited ability led to the development of a single day-trading strategy (SDTS) in which trading decisions are based on GP-forecasts of daily highest and lowest stock prices. SDTS executed for fifty consecutive trading days of six stocks yielded relatively high returns on investment.

Stock Returns in Thinly Traded Markets

AbstractWe examine the share price behavior of thinly traded NASDAQ National Market System stocks during periods when financial markets are open but the individual stocks do not trade. The absence of trade allows us to isolate the effect of nontrading from that of market closure. We find that nontrading stocks have negative mean returns and lower variances regardless of whether markets are open or closed. Two‐day returns that include one nontrading day have a mean daily return of ‐0.226% compared to +0.164% for two‐day returns over consecutive trading days. Two‐day returns that include one nontrading day have only 3.8% higher variance than one‐day returns. We conclude that the relation between transaction arrival, mean returns, and volatility depends on whether a stock is trading and not simply on whether the market is open.

The Birth of the Climax-Breadth Method

In addition to an immediate practical use for these figures, still retain a faintly nostalgic interest in this routine daily reporting of what have always called the of the market. My interest traces back to the fact that the first count of the market ever saw had been compiled by the late General Leonard P. Ayres. One morning in January 1926, General Ayres gave me a sheet of paper from a yellow pad on which there were several columns of figures. As he handed the paper over, he said, I think this will make an stock market statistic. It developed that the general had spent part of the previous evening counting the number of stock issues that had advanced, declined, and remained unchanged in each session of five consecutive trading days. My first reaction was one of more than casual interest although at that moment did not have the slightest idea that was going to devote thousands of hours over the next twenty-five years to this interesting stock market statistic. One of the first things thought about this new statistic was that the total number of issues traded would provide an exact measurement each day of the breadth of trading on the New York Stock Exchange. Knowing the total number of issues listed on the Exchange, we would have a daily figure showing the percentage of the listed issues that actually participated in the trading.