NASDAQ Market Research Articles

Financial market forecasting using a two-step kernel learning method for the support vector regression

In this paper, we propose a two-step kernel learning method based on the support vector regression (SVR) for financial time series forecasting. Given a number of candidate kernels, our method learns a sparse linear combination of these kernels so that the resulting kernel can be used to predict well on future data. The L 1-norm regularization approach is used to achieve kernel learning. Since the regularization parameter must be carefully selected, to facilitate parameter tuning, we develop an efficient solution path algorithm that solves the optimal solutions for all possible values of the regularization parameter. Our kernel learning method has been applied to forecast the S&P500 and the NASDAQ market indices and showed promising results.

The Role of the Informational Efficiency in the Dotcom Bubble

The financial bubbles are a challenge for the efficient market hypothesis. The phenomenon is always the same, a rapid increase in prices arriving to a maximum and then a strong and sudden decrease. The DotCom bubble is an example of the last century, the Nasdaq index closed at 5,048.62 on March 10, 2000, whereas it was 1,114 on August 1996, and also on October 2002. In this paper, we measure the informational efficiency of the Nasdaq market through the time using the Shannon entropy. Bootstrap simulations suggest that we can reject the hypothesis that the market behaved efficiently between August 17, 1998 and September 11, 2003. In addition, using a classical linear regression and a Tobit model, we present empirical evidence suggesting the existence of a negative relationship between the drawdown and the levels of efficiency. On the other hand, a Logit model presents some evidence of a negative relationship between the probability of crash and the informational efficiency. It suggests that a healthy market should present high level of efficiency. Finally, we propose a simply stochastic model simulating great part of the cluster of inefficiency detecting in the Nasdaq market.

Regulation and Bonding: The Sarbanes‐Oxley Act and the Flow of International Listings

ABSTRACTIn this paper, we examine the economic impact of the Sarbanes‐Oxley Act (SOX) by analyzing foreign listing behavior onto U.S. and U.K. stock exchanges before and after the enactment of SOX in 2002. Using a sample of all listing events onto U.S. and U.K. exchanges from 1995–2006, we develop an exchange choice model that captures firm‐level, industry‐level, exchange‐level, and country‐level listing incentives, and test whether these listing preferences changed following the enactment of SOX. After controlling for firm characteristics and other economic determinants of these firms' exchange choice, we find that the listing preferences of large foreign firms choosing between U.S. exchanges and the London Stock Exchange's (LSE) Main Market did not change following the enactment of SOX. In contrast, we find that the likelihood of a U.S. listing among small foreign firms choosing between the NASDAQ and LSE's Alternative Investment Market decreased following the enactment of SOX. The negative effect among small firms is consistent with these marginal companies being less able to absorb the incremental costs associated with SOX compliance. The screening of smaller firms with weaker governance attributes from U.S. exchanges is consistent with the heightened governance costs imposed by SOX increasing the bonding‐related benefits of a U.S. listing.

NASDAQ Velocity and Forces: An Interactive Visualization of Activity and Change

NASDAQ Market Velocity and Market Forces are two relatively new data products that attempt to capture market sentiment, something that was previously only observable if one was on a trading floor. Given the transient and temporal properties of the data, we were challenged to create a visualization that would highlight the ever-changing qualities of Velocity and Forces. To that end, we developed FireStox, a web application that provides unified representation and filtering solutions to help market researchers observe the behavior of these metrics for one or many companies throughout the course of a trading day.

Visibility effects and timing in stock listing changes: Evidence from operating performance

We extend the timing motive hypothesis to account for the influence of improved visibility as a result of a stock listing change. The results are, in general, consistent with this proposition, especially for Nasdaq and AMEX firms that switch to the NYSE. The empirical findings support the positive relationship of the visibility effect with post-listing operating returns. The evidence is mixed for the timing motive. The evidence is also mixed for both timing and visibility in the case of Nasdaq firms that move to the AMEX. An important finding is that post-listing operating performance has improved after the 1996 Nasdaq market reforms that reduced the NYSE advantage of lower liquidity costs and, thus, made opportunistic timing less important in listing change decisions. In addition, some evidence points to an association between differences in expected visibility improvement due to variation in listing standards and the pattern of post-listing operating performance.

Do Dealers Infer Information from Order Flow?

I examine strategic behavior amongst dealers in the NASDAQ market to identify whether they infer information from order flow. In particular, I ask: (a) is there a lead quote-setting dealer in each security and if so, are the quotes and/or trades of this leader informative?, (b) do others making market in the same security free-ride informative quotes?, and (c) what are some information signals that followers infer to identify the leader? I find that: (a) there is typically a quote-setting leader for each security who could be either a market maker or an Electronic Communications Network, (b) the quotes of this leader are followed by others making market in the same security, and (c) the lead dealer is usually the one that spends the highest percentage of time on the inside market, i.e., posts inside quotes. Trade volume is another (albeit weaker) information signal. My findings suggest that directing trade to the leader may be more advantageous to investors than randomly choosing a dealer for trade execution.

DO DEALERS INFER INFORMATION FROM ORDER FLOW?

AbstractI examine strategic behavior among dealers in the NASDAQ market and document that there is a lead quote‐setting dealer in each security and that the quotes posted by this leader are informative. Other dealers free‐ride this information by following the lead quote‐setting dealer. The lead dealer can be identified by two information signals: (1) percentage of time spent on the inside market (i.e., posting inside quotes), and (2) trade volume transacted. Dealers that free‐ride the leader's quotes quickly update their posted quotes in the same direction as the leader's quote change. My findings suggest that directing trade to the lead dealer may be more advantageous than randomly routing trade.

Price Discovery across the Stock Index Futures and the ETF Markets: Intra-Day Evidence from the S&P 500, Nasdaq-100 and DJIA Indices

This paper investigates the intra-day price dynamics of the S&P 500, Nasdaq-100 and DJIA indices for the periods both before and after the Nasdaq market crash which occurred between March 2000 and March 2001. We explore the relative price efficiencies of the three indices in the spot, futures, E-mini futures and ETF markets, and find that a cointegrating relationship existed between the three indices during the period after the crash. This would seem to imply that in the aftermath of the crash, the three indices shared common macroeconomic fundamentals. We find that where there is some disturbance in the equilibrium relationship between the indices, the market which adjusts to retain equilibrium is the Nasdaq-100 market. In the long run, the S&P 500 index leads the other index contracts, a finding which is consistent with the trading cost hypothesis. Nevertheless, the Nasdaq-100 index retains short-run price leadership over both the S&P 500 and DJIA indices.

Why effective spreads on NASDAQ were higher than on the New York stock exchange in the 1990s

Abstract Two hypotheses have been advanced to explain why spreads on NASDAQ were substantially higher than those on the NYSE in the 1990s: “collusion” and “preferencing and payment for order flow.” We present data on all actively traded stocks in these markets of relative effective spreads (RES), aggregated monthly over 1987–1999 and advance a third hypothesis: NASDAQ “SOES-day-trading.” We estimate NASDAQ and NYSE informed-trade losses and gains to market makers and other liquidity providers on six trade sizes, and find that losses on trades we ascribe to SOES day traders were substantially greater than those on other trades, offset somewhat by gains from small-trade-size investors. NASDAQ market makers' response to these losses and additional operations costs incurred to reduce the losses resulted in greater RES and increased trading within the best quotes, predominantly on larger trade sizes. The data are consistent with the “SOES-day-trading” hypotheses, but not with the other two. Furthermore, the mandatory SOES “experiment” provides insights into the negative effects of automated trading systems (such as ECNs, which now dominate NASDAQ) when their design does not adequately consider opportunistic traders.

Opinion Divergence and Post-Earnings Announcement Drift

This paper examines the relationship between divergent opinions and post-earnings announcement drift. We provide an improved measure of opinion divergence constructed from the dispersion of order flow across Nasdaq market makers that captures the breadth of divergence that is lost by volume-based measures. We find evidence that both limited participation (in the form of delayed price reaction and short sale constraints) and divergent opinions contribute significantly to drift. We also find that earnings surprises induce permanent upward shifts in opinion divergence, trading volume, and return volatility that last up to nine months following the announcement. Our results suggest that opinion divergence elicits added risk in the form of increased volatility with the resulting returns comprising a component of drift. We document that daily opinion divergence is a priced risk factor over the nine month drift period. The persistence of these relationships suggests that opinion divergence represents a fundamental change in the market's assessment of the announcing firm that extends beyond the announcement period and influences post-announcement stock returns.

Does Prepayment Risk of Mortgages Affect Excess Returns of Bank Stocks?

This paper explores the relationship between the prepayment risk embedded in conventional, fixed-rate residential mortgages and excess returns for bank stocks. There are two interesting findings in this study. First, commercial banks traded in the Nasdaq market are more meanvariance efficient than the other seven groups of industrial stocks. Second, the prepayment risk factor is significant for these banks. The prepayment risk mainly reflects a call option embedded in a mortgage plus foreclosure costs associated with a mortgage put option. This risk is measured by a remaining part of mortgage rates after excluding the influence of real estate market, maturity, and default risks on mortgage rates. The results of this study suggest that the prepayment risk factor does significantly affect excess returns for bank stocks in the period with high levels of mortgage refinancing activities.

The premium paid for M&A: The Nasdaq case

Our study is focused on the premium paid for an acquisition of a target company, especially on the Nasdaq market. We find that the relative size of the companies, the strategy of international diversification and the mean of payment influence the premium. There is no effect of maket timing on the premuim paid and the ownership structure of the group of directors doesn’t seem to be significant

Did NASDAQ Market Makers Successfully Collude to Increase Spreads? A Reexamination of Evidence from Stocks that Moved from NASDAQ to the New York or American Stock Exchanges

This paper examines all movements of stock from NASDAQ to the NYSE or Amex from 1983 through 1997 (1044 observations), as did Barclay (1997) for 472 stocks through 1992. He found a greater average decrease in spreads for NASDAQ stocks quoted on even eighths compared to stocks quoted on mixed eighths, from which he concluded that NASDAQ market makers had successfully colluded to widen spreads. A closer examination of the data and consideration of an important difference between how quotes are set on NASDAQ compared to the NYSE/Amex finds no support for the collusion hypothesis. Spreads on stocks with quoted spreads of one-eighth have no or limited scope for change, as this was the minimum quoted tick size on NYSE/Amex as well as on NASDAQ; hence inclusion of these observations reduces the average post-move spread decrease of mixed stocks, thereby biasing the finding. In fact, the higher average decrease in percentage effective spreads (%EPS) for even-eighth-quoted stocks is due entirely to the few stocks with high spreads (twelve eighths or higher). These stocks are relatively thinly traded and have small capitalizations and, when traded on NASDAQ (but not the NYSE/Amex) are subject to informed (adverse) trading. Furthermore, in the period after the alleged collusion was publicly identified and, presumably, stopped, there was no change in post-move decrease in %EPS on eveneighth stocks that supports the collusion hypotheses.

Why Effective Spreads on Nasdaq were Higher than on the New York Stock Exchange in the 1990s

Two hypotheses have been advanced to explain why spreads on Nasdaq were substantially higher than those on the NYSE in the 1990s - collusion and preferencing and payment for order flow. We present data on all actively traded stocks of relative effective spreads (RES) on these markets, aggregated monthly over 1987-1999 and advance a third hypothesis: Nasdaq SOES-day-trading. We estimate Nasdaq and NYSE informed-trade losses and gains to market makers on six trade sizes, and find that losses on trades we ascribe to SOES day traders were substantially greater than those on other trades, offset somewhat by gains to small-trade-size investors. Nasdaq market makers' response resulted in greater RES and increased trading within the best quotes, predominantly on larger trade sizes. Among market makers' customers, the costs of SOES day trading were borne largely by traders of the smaller trade sizes. The data are consistent with the SOES-day-trading hypotheses, but not with the other two. Furthermore, the mandatory SOES experiment gave impetus to trading via ECNs, which now dominate Nasdaq.

Generalized regression model for sequence matching and clustering

Linear relation has been found to be valuable in rule discovery of stocks, such as if stock X goes up a, stock Y will go down b. The traditional linear regression models the linear relation of two sequences faithfully. However, if a user requires clustering of stocks into groups where sequences have high linearity or similarity with each other, it is prohibitively expensive to compare sequences one by one. In this paper, we present generalized regression model (GRM) to match the linearity of multiple sequences at a time. GRM also gives strong heuristic support for graceful and efficient clustering. The experiments on the stocks in the NASDAQ market mined interesting clusters of stock trends efficiently.

Current NASDAQ Corporation Methods of Reporting Comprehensive Income

Statement of Financial Accounting Standard No. 130 (SFAS 130) was released in 1997 which required publicly traded companies to separately report comprehensive income in the financial statements. SFAS 130 prescribed three alternative formats for the presentation without mandating any one specific format. SFAS 130 also required certain details of comprehensive income to be displayed prominently in the financial statements. The current study examined the presentation of comprehensive income by a sample of companies traded on the NASDAQ market to determine the predominant method of presentation among these companies, five years after SFAS 130 became effective. Results of the examination of one hundred annual reports showed that a majority of the sampled companies used the third approach, which was to present comprehensive income as part of the Statement of Stockholders’ Equity, as against the first two approaches that favored presentation of comprehensive income on the face of the Income Statement or in a separate statement. Further, the paper also discusses some ancillary findings pertaining to the presentation of the details of comprehensive income.

Comparing the stock recommendation performance of investment banks and independent research firms

From January 1996 through June 2003, the average daily abnormal return to independent research firm buy recommendations exceeds that of investment bank buy recommendations by 3.1 basis points (almost 8 percentage points annualized). Investment bank buy recommendation underperformance is more pronounced following the NASDAQ market peak (March 10, 2000) and strikingly so for buy recommendations on firms that recently conducted equity offerings. In contrast, investment bank hold and sell recommendations outperform those of independent research firms by 1.8 basis points daily ( 4 1 2 percentage points annualized). These results suggest reluctance by investment banks to downgrade stocks whose prospects dimmed during the bear market of the early 2000s, as claimed in the SEC's Global Research Analyst Settlement.

Investor Irrationality and the Nasdaq Bubble

We exploit the information in the options market to study the risk and risk premium variations around the Nasdaq bubble period. In particular, we investigate whether the dramatic rise and fall of the Nasdaq can be justified by changes in return risk, or there were corresponding unusual shifts in how investors price risks. We specify a model that accommodates fluctuations in both risk levels and market prices of different sources of risks, and we estimate the model using time-series returns and option prices on the Nasdaq 100 index. Our analysis reveals three key pieces of evidence that foretell the arrival and burst of a bubble. First, during the Nasdaq bubble period, return volatility increased together with the rising index level, even though the two tend to move in opposite directions in general. Second, while the market price of volatility risk is strongly negative historically as investors dislike high volatility levels and volatility risk, the market price of volatility risk declined in absolute magnitude and approached zero at the end of 1999. In contrast to the average risk premium at normal market conditions, aversion to volatility risk completely disappeared during the height of the bubble period. Third, the options market showed an increasing market price of downside jump risk that peaked three month prior to the bursting of the bubble, highlighting the increasing demand for hedging against the potential crash of the Nasdaq market valuation. On the other hand, the market price of the downside jump risk dropped suddenly three months prior to the burst, pointing to capital flight from the Nasdaq market and the unwinding of the associated option hedging positions. Our approach suggests that having options on the bubble asset provides a window into investor behavior unavailable from earlier bubbles.

A Comparison of Earnings Management Between KSE Firms and KOSDAQ Firms

Abstract: This study investigates differences in earnings management practices of Korea Stock Exchange (KSE) firms and KOSDAQ (a Korean version of the NASDAQ market) firms during the period of 1996–1997. A sample of 1,256 KSE and 577 KOSDAQ firm‐year observations is used to compare earnings management practices of firms listed in the two different stock exchanges. The results of the study reveal that KOSDAQ firms tend to more actively manipulate earnings to avoid losses than KSE firms. KOSDAQ firms generally tend to increase reported earnings more aggressively than KSE firms when their operating cash flows are poor, and play down their reported earnings more when their operating cash flows are exceptionally good. The results of the study are quite robust in the sense that more aggressive earnings management practices of KOSDAQ firms persist even when operating cash flows are controlled.

A micro-movement model with Bayes estimation via filtering: Application to measuring trading noises and costs

We highlight a general hybrid system as the micromovement model for asset price using counting processes recently introduced with its Bayes estimation via filtering. We construct a new simple micromovement model and apply it to analyze trade-by-trade stock price data in the light of the series of works initiated by Christie and Schultz [Why do NASDAQ market makers avoid odd-eighth quotes?, Finance 49 (1994) 1813–1840]. Through the new model, we propose more reasonable, but computationally intensive measures for trading noise including clustering noise and non-clustering noise, and for trading cost. We employ Bayes estimation via filtering to obtain parameter estimates of the new model and to provide numerical measures of trading noise and trading cost for three stocks from four chosen periods. Our empirical results support the important findings in [Christie, Harris, Schultz, Why did NASDAQ market makers stop avoiding odd-eighth quotes?, Finance 49 (1994) 1841–1860; Barclay, Christie, Harris, Kandel, Schultz, The effects of market reform on the trading costs and depths of NASDAQ stocks, J. Finance 54(1) (1999) 1–34].