Andrew C. Pollock

An application of probability judgement accuracy measures to currency forecasting

Abstract A general framework for examining the quality of judgement in a currency forecasting context is described. This framework incorporates existing probability judgement accuracy measures and adapts them in a way that takes full advantage of the precise nature of exchange rate time series data. It also includes error measures, which are widely used in time series contexts, and adapts them to a probabilistic form in order to provide additional information in specific circumstances. The advantages of the proposed measures are outlined and an application to some actual currency data is provided. The framework is described initially to deal with directional forecasts and then extended to deal with point value, statistical and composite forecasts.

Feedback-labelling synergies in judgmental stock price forecasting

Research has suggested that outcome feedback is less effective than other forms of feedback in promoting learning by users of decision support systems. However, if circumstances can be identified where the effectiveness of outcome feedback can be improved, this offers considerable advantages, given its lower computational demands, ease of understanding and immediacy. An experiment in stock price forecasting was used to compare the effectiveness of outcome and performance feedback: (i) when different forms of probability forecast were required, and (ii) with and without the presence of contextual information provided as labels. For interval forecasts, the effectiveness of outcome feedback came close to that of performance feedback, as long as labels were provided. For directional probability forecasts, outcome feedback was not effective, even if labels were supplied. Implications are discussed and future research directions are suggested.

The quality of bank forecasts: The dollar-pound exchange rate, 1990–1993

Abstract A general framework for examining the quality of currency forecasts is described. This framework incorporates existing accuracy measures and modifies them to give components of accuracy that are appropriate for statistically derived and judgementally based forecasts. The framework is described and applied to one week ahead US

The influence of the forecast horizon on judgemental probability forecasts of exchange rate movements

/UK£ forecasts from three major banks over a three year period between 1990 and 1993. The results suggest that, while overall forecast performance was poor, some aspects of the predictions could still be useful in a practical setting.

Directional Judgemental Financial Forecasting: Trends and Random Walks

An experiment is reported which compares directional forecasting performance of experts, novices and simple statistical models over three time horizons on a task involving probabilistic forecasts of exchange rate movements. Probability-judgement accuracy analyses illustrated no clear overall performance differences between experts and novices, but significant differences between the groups on various important components of judgement suggested that the groups obtained their similar overall scores using different cognitive strategies. Striking horizon effects and expertize–horizon interactions were also observed. The subjects performed better than a random walk forecaster, but worse than the random walk with constant drift and first-order autoregressive models. Composite human judgement, however, not only improved on individual judgement but, also, surpassed the simple statistical models in many instances. Possible explanations are offered for these results, suggestions are made for future research, and practical implications are emphasized.

Currency Forecasting: An Investigation Into Probability Judgement Accuracy

In the volatile and rapidly changing financial environment human judgement is an essential ingredient in decision-making. Effective exchange rate and stock price forecasting not only require quantitative models but also the personal beliefs and views of a forecaster formed according to some subjective procedure. It is important, then, to understand how people involved in the financial world make judgements on currency or stock price movements. This paper reports an exploratory investigation of this issue undertaken with a sample of delegates at the 9th. Meeting of the Euro-Working Group on Financial Modelling at Curacao, Netherlands Antilles, in April 1991.

Evaluating strategic directional probability predictions of exchange rates

Judgemental forecasting techniques are extensively used in business and industry (eg. Dalrymple, 1987; Klein and Linneman, 1984). Currency forecasting is no exception. Professionals in the currency market often rely on judgement alone or, at the very least, incorporate some sort of subjective procedure into their forecasts. Despite this, however, academic research has tended to focus on the behaviour of statistical models and has neglected what appears to be the most important aspect of real-world forecasting: human judgement.

Can Subjective Exchange Rate Forecasts be Improved

The current paper aims to examine strategic predictions (with forecast horizons greater than six months) via the empirical probability (EP) technique. This technique was proposed initially to examine short-term tactical predictions (with forecast horizons less than three months), as set out in Pollock et al. (2005). The proposed procedure is based on the hypothesis that changes in logarithms of daily exchange rates follow a normal distribution over short horizons (of 10 to 30 days), but longer term forecast evaluation requires consideration of cumulative parameters consistent with changing means and standard deviations arising from primary and secondary trends. It is shown that ex-post EPs can be obtained for any predictive horizon above 30 days (e.g., 180 days) by using a combination of shorter (e.g., 20-day) Student t distributions. The procedure is illustrated using daily Euro/USD series from 4 January 1999 to 29 January 2008 to evaluate a set of Euro/USD directional probability predictions.

The relative influence of advice from human experts and statistical methods on forecast adjustments

In the volatile and rapidly changing financial environment, human judgement is essential to currency forecasting. Recent events like German reunification, Yugoslavian upheaval, Iraq’s invasion of Kuwait and the breakup of the USSR are a striking reminder that quantitative models alone cannot predict the future. Reliable currency forecasting requires subjective input, using the expertise of the person or persons making the forecast. Ignoring this knowledge can be a serious handicap: exchange rate movements are often influenced by unique events that can often be broadly foreseen, but cannot easily be incorporated into conventional models. People have many advantages over quantitative models, especially flexibility and their ability to take the unexpected into account.