Search results

The existence of weak‐form efficiency in the equity markets of the three main Central European transition economies (the Czech Republic, Hungary, and Poland) is examined for the period July 1995 through September 2000, using weekly Investable and Comprehensive indexes developed by the International Finance Corporation. Several different approaches are used. Univariate and multivariate tests provide some evidence that stock prices in these exchanges exhibit a random walk, which constitutes evidence for weakform efficiency. This differs in some cases from studies using data for the initial years of these markets. The variance ratio test (VR) of Lo and MacKinlay (1988) yields somewhat mixed results concerning the random‐walk properties of the indexes. A modelcomparison test compares forecasts from a NAÏVE model with ARIMA and GARCH alternatives. Results from the model‐comparison approach are consistent in rejecting the random‐walk hypothesis for the three Central European equity markets.

This chapter introduces the best linear predictor (BLP) with the asymptotic minimum mean squared forecasting error (MSFE) among linear predictors of variables in cointegrated systems. Accordingly, the authors show that (i) if the autocorrelation coefficient of the cointegration error between the prediction time and the predicted targeting time is larger than ½ (representing a short prediction period), then the BLP is deduced from the random walk model; and (ii) in other cases (representing a long prediction period), the BLP is deduced from the cointegration model. Under this scheme, we suggest a switching predictor that automatically selects the random walk or cointegration model according to the size of the estimated autocorrelation coefficient. These results effectively explain the superiority reversal in the short- and long-term prediction of the exchange rate between the random walk and the structural/cointegration model (known as the Meese–Rogoff or disconnect puzzle).

To develop an integrated approach to forecasting spot foreign exchange rates by incorporating some principles underlying long‐term dependence.

The paper utilises the random‐walk framework to develop a stochastic forecast model wherein the sign (positive or negative) and magnitude (strong or weak) of dependence can be separately controlled. The integrated model demonstrates superior forecast performance over a conventional random walk.

Using spot log prices and log price changes (returns) for the USD/AUD exchange rate, the initial outcomes of the study suggest that a priori knowledge of the underlying sign and magnitude of long‐term dependence yields out‐of‐sample forecasts superior to those of a random walk model.

Independent assessment of the contribution to forecast accuracy of controlling for the sign of dependence between successive price changes only shows little additional improvement in out‐of‐sample forecast performance over the random walk null.

The findings of the study have important ramifications for managerial finance as they provide important insights on expected future currency returns with potential advantages in currency hedging and/or timing of international capital flows.

The contribution of this paper is to develop an original forecast model explicitly incorporating the conceptual and theoretical characteristics of long‐term dependent time series. By separating the key characteristics and modelling each individually, the contribution of each to forecast accuracy can be evaluated.

Product success is contingent on forecasting when a product is needed and how it should be offered. Forecasting accuracy is contingent on the correct forecasting technique. Using supermarket data across two product categories, this chapter shows that using a bevy of forecasting methods improves forecasting accuracy. Accuracy is measured by the mean absolute percentage error. The optimal methods for one consumer goods product may be different than for another. The best model varied from sophisticated, most such as autoregressive integrated moving average (ARIMA) and Holt–Winters to a random walk model. Forecasters must be proficient in multiple statistical techniques since the best technique varies within a categories, variety, and product size.

This study seeks to measure the behaviour of stock prices in the Bahrain Stock Exchange (BSE), which is expected to follow a random walk. The aim of the study is to measure the weak‐form efficiency.

Random walk models such as unit root and Dickey‐Fuller tests are used as basic stochastic tests for a non‐stationarity of the daily prices for all the listed companies in the BSE. In addition, autoregressive integrated moving average (ARIMA) and exponential smoothing methods are also used. Cross‐sectional‐time‐series is used for the 40 listed companies over the period 1 June 1990 up until 31 December 2000.

Random walk with no drift and trend is confirmed for all daily stock prices and each individual sector. Other tests, such as ARIMA (AR1), autocorrelation tests and exponential smoothing tests also supported the efficiency of the BSE in the weak‐form.

The finding of the study is a necessary piece of information for all investors whether in Bahrain or dealing with Bahrain stock market. Listed firms could also benefit from the findings by seeing the true picture of their stock price. Since, Bahrain is considered as an emerging market, the new methodologies used could be replicated for all other emerging markets. In addition, the finding is used as a base for testing the market efficiency in the semi‐strong form, which has not yet been tested by any researcher.

This study will add value to the literature of market efficiency in emerging market since it is the only study which covers all the listed companies and over a long period of time. To confirm the weak‐form efficiency in Bahrain, the study is unique in using five different methods in the same paper which have not been found in the previous literature.

The random walk forecast of exchange rate serves as a standard benchmark for forecast comparison. The purpose of this paper is to assess whether this benchmark is unbiased and directionally accurate under symmetric loss. The focus is on the random walk forecasts of the dollar/euro for 1999‐2007 and the dollar/pound for 1971‐2007.

A forecasting framework to generate the one‐ to four‐quarter‐ahead random walk forecasts at varying lead times is designed. This allows to compare forecast accuracy at different lead times and forecast horizons. Using standard evaluation methods, this paper further evaluates these forecasts in terms of unbiasedness and directional accuracy.

The paper shows that forecast accuracy improves with a reduction in the lead time but deteriorates with an increase in the forecast horizon. More importantly, the random walk forecasts are unbiased and accurately predict directional change under symmetric loss and thus are of value to a user who assigns similar cost to incorrect upward and downward move predictions in the exchange rates.

The one‐ to four‐quarter‐ahead random walk forecasts evaluated here are for averages of daily figures and not for the (end‐of‐quarter) rates in 3‐, 6‐, 9‐ and 12‐months. Thus, the framework is of value to a market participant who is interested in forecasting quarterly average rates rather than the end‐of‐quarter rates.

The exchange rate forecasting framework presented in this paper allows the evaluation of the random walk forecasts in terms of directional accuracy which (to the best of knowledge) has not been done before.

This paper examines the Random Walk Hypothesis (RWH) for aggregate New Zealand share market returns, as well as the CRSP NYSE‐AMEX (USA) index during the 1980‐2001 period. Using several indices, we rely on the variance‐ratio test and find evidence to support the rejection of the RWH with some evidence of a momentum effect. However, we find evidence to suggest the behaviour of share prices to be time‐dependent in New Zealand. For example, we find the indices tested were closer to random after the 1987 share market crash. Further analysis showed even stronger results for periods subsequent to the passage of the Companies Act 1993 and the Financial Reporting Act 1993. We also find evidence that indices based on large capitalisation stocks are more likely to follow a random walk compared to those based on smaller stocks. For the USA index, we find stronger evidence of random behaviour in our sample period compared to the earlier period examined by Lo and Mackinlay (1988)

Several lessons learnt from a Bayesian analysis of basic macroeconomic time-series models are presented for the situation where some model parameters have substantial posterior probability near the boundary of the parameter region. This feature refers to near-instability within dynamic models, to forecasting with near-random walk models and to clustering of several economic series in a small number of groups within a data panel. Two canonical models are used: a linear regression model with autocorrelation and a simple variance components model. Several well-known time-series models like unit root and error correction models and further state space and panel data models are shown to be simple generalizations of these two canonical models for the purpose of posterior inference. A Bayesian model averaging procedure is presented in order to deal with models with substantial probability both near and at the boundary of the parameter region. Analytical, graphical, and empirical results using U.S. macroeconomic data, in particular on GDP growth, are presented.

This paper examines the opportunity cost of applying simple averages in formulating the Department of Defense (DoD) budget for foreign exchange rates. Using out-of-sample validation, we evaluate the status quo of a center-weighted average against a Random Walk model, ARIMA, forward rates, futures contracts, and a private firm's forecasts over two time periods extending from Fiscal Year (FY) 1991 to FY 2014. The results strongly indicate that four of the alternative methods outperform the status quo over the shorter time period, and three methods for both time periods. Furthermore, a non-parametric comparison of the median error demonstrates statistical similarities between the four alternative methods over the short term. Overall, the paper recommends using the futures option prices to decrease forecast error by 3.23% and avoiding a $34 million opportunity cost.