Numerical issues in threshold autoregressive modeling of time series

This paper analyses the contribution of various numerical approaches to making the estimation of threshold autoregressive time series more efficient. It relies on the computational advantages of QR factorizations and proposes Givens transformations to update these factors for sequential LS problems. By showing that the residual sum of squares is a continuous rational function over threshold intervals it develops a new fitting method based on rational interpolation and the standard necessary optimality condition. Taking as benchmark a simple grid search, the paper illustrates via Monte Carlo simulations the efficiency gains of the proposed tools.     

Long memory and structural breaks in commodity futures markets

This study employs daily data for 14 commodities and three financial assets 1990–2009 to explore the impact of the time series properties of the futures‐spot basis and the cost of carry on forward market unbiasedness. The main result is that the basis of 16 assets exhibits both long memory and structural breaks. The long memory in the basis is robust even to the use of break‐adjusted data. It implies that the cost‐of‐carry has long memory which the empirical results confirm using the interest cost as a proxy. These new findings suggest that the forecast error has long memory and are inconsistent with unbiasedness. They could be consistent with a weaker version of market efficiency in the presence of a fractionally integrated, time‐varying risk premium but they could also be rationalized by priced noise trader risk with limits to arbitrage in less than fully efficient markets. © 2010 Wiley Periodicals, Inc. Jrl Fut Mark     

Artificial Neural Networks Applied to Ratio Analysis in the Analytical Review Process

Experts claim that artificial neural network (ANN) technology can outperform standard statistical methods when applied to examine actual financial data. Researchers have used ANNs to analyze bankruptcy prediction, bond rating and the going-concern problem. Financial firms have employed ANNs commercially to predict commercial bank failures, detect credit card fraud and verify signatures. For accounting and auditing problems, however, application of ANN technology has been limited. Preliminary experiments tested whether an ANN offered improved performance in recognizing material misstatements during the analytical review process of auditing. Four years of audited financial data from a medium-sized distributor were input as data streams to calibrate the ANN across fifteen financial accounts. Researchers compared a presumed lack of actual errors and certain seeded material errors with signals from the ANN analytical review process to evaluate performance. Results were compared to analyses where financial ratios and regression methods were employed as analytical review techniques. Results tentatively suggest that the ANN method recognized patterns within financial accounts more effectively than did financial ratio and regression methods. ANNs applied as a forecasting tool seem useful for identifying patterns that can indicate potential investigations of a firm's unaudited financial data in the current year.     

Artificial neural networks in accounting and finance: modeling issues

This article reviews the literature on artificial neural networks (ANNs) applied to accounting and finance problems and summarizes the ‘suggestions’ from this literature. The first section reviews the basic foundation of ANNs to provide a common basis for further elaboration and suggests criteria that should be used to determine whether the use of an ANN is appropriate. The second section of the paper discusses development of ANN models including: selection of the learning algorithm, choice of the error and transfer functions, specification of the architecture, preparation of the data to match the architecture, and training of the network The final section presents some general guidelines and a brief summary of research progress and open research questions. Copyright © 2000 John Wiley & Sons, Ltd.     

The Fourth International Symposium on Intelligent Systems in Accounting,Finance and Management

The Fourth International Symposium on Intelligent Systems in Accounting, Finance and Management ranks as the oldest, on-going meeting on the use of artificial intelligence (AI) in business. The conference spotlights a wide range of issues with both theoretical/practical and behavioral/technological dimensions.     

Commodity futures returns: more memory than you might think!†

This paper investigates long-range dependence in 14 commodity and 3 other financial futures returns series from 1993 to 2009 and shows that long memory is a pervasive phenomenon in contrast to the extant evidence. Utilizing a semi-parametric wavelet-based estimator with time windows, the results provide overwhelming evidence of time-varying long-range dependence in all futures returns series. Structural break tests indicate multiple regimes of dependence, in the majority of which the persistence parameter is statistically significant. The results also provide evidence of predominantly negative parameter values which are known as anti-persistence. The latter is consistent with investor overreaction to shocks and suggests temporary departures from market efficiency.     

Does Volatility Improve UK Earnings Forecasts?

Abstract:  We investigate the relation between UK accounting earnings volatility and the level of future earnings using a unique sample comprising some 10,480 firm-year observations for 1,481 non-financial firms over the 1985–2003 period. The findings confirm the in-sample result of an inverse volatility-earnings relation only for the 1998–2003 sub-period and for the most profitable firms. The out-of-sample forecast accuracy for the top earnings quintile improves when volatility is added as a regressor to a model including only lagged earnings. The findings are consistent with the over-investment hypothesis and the view that the earnings of the most volatile firms tend to mean revert more rapidly.     

Post-IPO Operating Performance, Venture Capital and the Bubble Years

Abstract:  We analyse the post-issue operating performance of 316 venture-backed and 274 non-venture UK IPOs 1985–2003. The finding of a statistically significant five-year, operational decline exhibited over the full sample period is not robust. Rather it is driven by the dramatic underperformance during the 1998–2000 bubble years while IPOs perform normally in the remaining years. Cross-section regression results indicate support for venture capital certification in the non-bubble years but a significantly negative relationship between operating performance and venture capitalist board representation during the bubble years. The bubble year underperformance is explained by market timing and by low quality companies taking advantage of investor sentiment.     

Earnings management and IPO anomalies in China

This paper empirically examines the impact of earnings management and investor sentiment on IPO anomalies using a sample of 506 Chinese IPOs issued over the 1998–2003 period. We develop a parsimonious pricing model in which both the offer price and the short-term aftermarket price are influenced by the use of earnings management, and show that the offer price can be below the fair price while the short-term equilibrium price in the aftermarket can be overvalued due to investor sentiment. Consistent with the overreaction hypothesis, the empirical results reveal a positive relation between the initial return and managed accruals and a negative relation between the long-term stock performance and the initial return. Earnings management appears to generate a pattern where the initial price following an IPO tends to be inflated by overreaction in the secondary market but adjusts to its fundamental level in the long run. These findings are robust across a variety of test specifications.