An explainable machine learning approach to improve take-off time predictions

Accurate aircraft trajectory predictions are necessary to compute exact traffic demand figures, which are crucial for an efficient and effective air traffic flow and capacity management. At present, the uncertainty of the take-off time is one of the major contributions to the loss of trajectory predictability. In the EUROCONTROL Maastricht Upper Area Control Centre, the predicted take-off time for each individual flight relies on the information received from the Enhanced Traffic Flow Management System. However, aircraft do not always take-off at the times reported by this system due to several factors, which effects and interactions are too complex to be expressed with hard-coded rules. Previous work proposed a machine learning model that, based on historical data, was able to predict the take-off time of individual flights from a set of input features that effectively captures some of these elements. The model demonstrated to reduce by 30% the take-off time prediction errors of the current system one hour before the time that flight is scheduled to depart from the parking position. This paper presents an extension of the model, which overcomes this look-ahead time constraint and allows to improve take-off time predictions as early as the initial flight plan is received. In addition, a subset of the original set of input features has been meticulously selected to facilitate the implementation of the solution in an operational air traffic flow and capacity management system, while minimising the loss of predictive power. Finally, the importance and interactions of the input features are thoroughly analysed with additive feature attribution methods.     

The impact of privatization on efficiency and profitability

Over the past three decades, China has undergone tremendous economic and social change as a consequence of the transition from a centrally planned to a market economy. This paper examines a key feature of this transition – the privatization of the state‐owned enterprises (SOEs) – through both a theoretical model and empirical analysis. Using newly collected primary data from a variety of sources, we study how privatization of listed SOEs affects employment, wages, profits and other aspects of economic performance at the firm level. Our major finding is that privatization results in substantial downsizing of employment, increased labour productivity and rising profitability.     

A NOTE ON PRICING OF PRODUCT QUALITY FOR STATUS CONCERNS

The paper analyses the optimal pricing of the product quality scheme when concerns for relative standing exist among consumers. We demonstrate that if the proportion of high‐value consumers is over (respectively, under) 1/2 of the total consumers, a firm has an incentive to select a large (respectively, small) quality gap among products. Therefore, there exists a cut‐off level for status concerns, which eliminates quality differences, and the firm assigns the same quality to all the consumers. These results indicate that consumers’ qualities will reflect distortions at the top and bottom. Accordingly, the firm's profit depends on which consumer category is larger.     

Tourists’ impulse buying behavior at duty-free shops: the moderating effects of time pressure and shopping involvement

This study investigates the consumer behavior of tourists at duty-free shops who experience different limitations on shopping than do consumers at general shopping malls. The target for sampling was outbound tourists who shopped in two hotel duty-free shops in Seoul, Korea. The final analysis used data from 343 respondents and the results revealed (1) the more positive the emotions felt by tourists, the more impulsive buying behavior occurs; (2) time pressure during shopping reinforces negative emotions, resulting in increased affective impulse buying; and (3) increased shopping involvement directly increases cognitive impulse buying.     

TIME‐CONSISTENT AND MARKET‐CONSISTENT EVALUATIONS

We consider evaluation methods for payoffs with an inherent financial risk as encountered for instance for portfolios held by pension funds and insurance companies. Pricing such payoffs in a way consistent to market prices typically involves combining actuarial techniques with methods from mathematical finance. We propose to extend standard actuarial principles by a new market‐consistent evaluation procedure which we call “two‐step market evaluation.” This procedure preserves the structure of standard evaluation techniques and has many other appealing properties. We give a complete axiomatic characterization for two‐step market evaluations. We show further that in a dynamic setting with continuous stock prices every evaluation which is time‐consistent and market‐consistent is a two‐step market evaluation. We also give characterization results and examples in terms of g‐expectations in a Brownian‐Poisson setting.     

Tax avoidance and ex ante cost of capital

This study examines the association between tax avoidance and ex ante cost of equity capital. Based on prior research, we develop two proxies for investors’ expectations of tax avoidance and explore whether deviations from those expectations result in higher ex ante cost of equity capital. We find that the ex ante cost of equity capital increases with tax avoidance that is either below or above investor expectations and that the increase is larger for tax avoidance that exceeds investors’ expectations. We then examine whether firms that alter their future tax avoidance exhibit a lowering of their ex ante cost of equity capital and find that tax avoidance decreases (increases) from the prior year for firms that were above (below) investors’ expectations in the prior year. These results are consistent with the trade‐off suggested by the Scholes and Wolfson framework and reinforce the notion that balancing tax benefits and non‐tax costs is an important feature of firms’ tax planning.     

The Unintended Consequences of the Frequency of PCAOB Inspection

After more than 50 years of self‐regulation of the US auditing profession, the Sarbanes‐Oxley Act of 2002 (SOX) created the Public Company Accounting Oversight Board (PCAOB) as a quasi‐governmental entity with statutory authority to inspect accounting firms that audit public clients. The frequency of this inspection is annual or triennial, based upon the number of public clients the firm audits. We examine the effects of these two levels of inspection frequency on financial reporting quality and audit fees for clients of small and midsize public accounting firms. Our findings provide evidence of significantly higher audit quality and audit fees for clients of annually inspected firms relative to clients of triennially inspected firms. These findings are robust to auditor‐client alignment analyses, propensity score matching, time‐series analyses, examination of firms that have changed from triennial to annual inspection, and particular examination of firms with inspection deficiencies. Overall, our study suggests that the two‐tier frequency system of PCAOB inspection may have also resulted in two‐tier audit quality and audit fee systems for small and midsize public accounting firms, with more frequent inspection leading to more rigorous and informed auditor decisions. We discuss the implications of our results for the Board and the profession at large.     

Simultaneous Estimation of Legal and Illegal Supply and Demand: The Case of Motion Pictures

Prior research into illegal goods has typically looked at one-way effects, such as illegal demand on legal demand. This research investigates a previously unexamined component of the market, illegal supply. The authors examine the supply and demand of legal goods and their illegal counterparts as a market system of four interdependent components. This research makes theoretical and empirical contributions by evaluating illegal supply in this system. Simultaneous equations estimate each market component on the others using data from the motion picture industry. The results find illegal supply has no effect on legal supply (movie screens), positive effects on illegal demand (piracy downloads), and some effect on legal demand (box office revenues). Timing effects highlight this: illegal supply has a positive effect on legal demand during a film’s opening week, but no effect post-launch. The other market components have positive effects on illegal supply (except legal supply, which is negative in the opening week). Additionally, illegal demand has a negative effect on legal demand during the opening week of release, but not in the subsequent weeks. This finding alleviates prior research tension as to whether piracy helps or hurts legal sales, as omitting illegal supply could result in biased estimates.     

A saddlepoint approximation to the distribution of the sum of independent non‐identically uniform random variables

Calculating the probability of the corresponding significance point is important for finite sample sizes. However, it is difficult to evaluate this probability when the sample sizes are moderate to large. Under these circumstances, consideration of a more accurate approximation for the distribution function is extremely important. Herein, we performed a saddlepoint approximation in the upper tails for the distribution of the sum of independent non‐identically uniform random variables under finite sample sizes. Saddlepoint approximation results were compared with those for a normal approximation. Additionally, the order of errors of the saddlepoint approximation was derived. © 2014 The Authors. Statistica Neerlandica © 2014 VVS.