Random Walk Forecasts Research Articles

High Dimensional Yield Curves: Models and Forecasting

Functional Signal plus Noise (FSN) models are proposed for analysing the dynamics of a large cross-section of yields or asset prices in which contemporaneous observations are functionally related. The FSN models are used to forecast high dimensional yield curves for US Treasury bonds at the one month ahead horizon. The models achieve large reductions in mean square forecast errors relative to a random walk for yields and readily dominate both the Diebold and Li (2006) and random walk forecasts across all maturities studied. We show that the Expectations Theory (ET) of the term structure completely determines the conditional mean of any zero-coupon yield curve. This enables a novel evaluation of the ET in which its 1-step ahead forecasts are compared with those of rival methods such as the FSN models, with the results strongly supporting the growing body of empirical evidence against the ET. Yield spreads do provide important information for forecasting the yield curve, especially in the case of shorter maturities, but not in the manner prescribed by the Expectations Theory.

Is there an empirical link between the dollar price of the euro and the monetary fundamentals?

This paper examines the empirical link between the dollar exchange rate of the euro and the monetary fundamentals. The exchange rate is found to be cointegrated with money and income differentials, while the homogeneity restrictions are supported by the data. The weak form restrictions of the present-value model of the foreign exchange market are not rejected by the data, but the most stringent restrictions are strongly rejected. An estimated error-correction model explains a substantial part of the short-run exchange rate volatility and outperforms the random walk forecasts.

Analyst Earnings Forecasts for Publicly Traded Insurance Companies

Several trends in the insurance and financial services industry, including demutualizationconsolidation, and deregulation, have attracted increasing attention from investors and financial analysts. This paper investigates the accuracy of the earnings forecasts of financial analysts for insurance companies. Our empirical results indicate that analyst forecasts outperform random walk time-series forecasts. Furthermore, we find that both disagreement over earnings forecasts among analysts and the relative forecasting error in the mean forecasts is smaller for life insurers than for property-casualty insurers, whereas the relative errors for forecasts for multiple-line insurers are in between the two. Forecasting error is a negative function of firm size and the number of analysts who are following a company, and is a positive function of the disagreement among analysts.Analyst forecasts have a timing advantage over the random walk model. Our results also suggest that the fair value reporting requirement (SFAS 115), which has been in effect since 1994, has enhanced the accuracy of analyst forecasts. The SFAS 115 has improved the superiority of analyst forecasts over the random walk forecasts for life insurers, but not for property-casualty insurers, and there is a weak improvement for multiple-line insurers.

EXPLORING THE IMPACT OF CALENDAR EFFECTS ON THE DYNAMIC STRUCTURE AND FORECASTS OF FINANCIAL TIME SERIES

Several recently developed chaotic forecasting methods give better results than the random walk forecasts. However they do not take into account specific regularities of stock returns reported in empirical finance literature, such as the calendar effects. In this paper, we present a method for filtering the day-of-the-week and the holiday effect in a time series. Our main objective is twofold. On the one hand we study how the underlying dynamics of the Nasdaq Composite, and TSE 300 Composite returns series can be influenced by the presence of calendar effects. On the other hand we adapt our method to chaotic forecasting. Its computational advantages lead to significant improvements of forecasts.

Using Forecasts of Forecasters to Forecast

Quantification techniques are popular methods in empirical research to aggregate the qualitative predictions at the micro-level into a single figure. In this paper, we analyze the forecasting performance of various methods that are based on the qualitative predictions of financial experts for major financial variables and macroeconomic aggregates. Based on the Centre of European Economic Research's Financial Markets Survey, a monthly qualitative survey of around 330 financial experts, we analyze the out-of-sample predictive quality of probability methods and regression methods. Using the modified Diebold-Mariano-Test of Harvey, Leybourne & Newbold (1997), we confront the forecasts based on survey methods with the forecasting performance of standard linear time series approaches and simple random walk forecasts.

ARE EXCHANGE RATES REALLY RANDOM WALKS? SOME EVIDENCE ROBUST TO PARAMETER INSTABILITY

Many authors have documented that it is challenging to explain exchange rate fluctuations with macroeconomic fundamentals: a random walk forecasts future exchange rates better than existing macroeconomic models. This paper applies newly developed tests for nested models that are robust to the presence of parameter instability. The empirical evidence shows that for some countries we can reject the hypothesis that exchange rates are random walks. This raises the possibility that economic models were previously rejected not because the fundamentals are completely unrelated to exchange rate fluctuations, but because the relationship is unstable over time and, thus, difficult to capture by Granger causality tests or by forecast comparisons. We also analyze forecasts that exploit the time variation in the parameters and find that, in some cases, they can improve over the random walk.

A Test for Superior Predictive Ability

We propose a new test for superior predictive ability. The new test compares favorably to the reality check (RC) for data snooping, because it is more powerful and less sensitive to poor and irrelevant alternatives. The improvements are achieved by two modifications of the RC. We use a studentized test statistic that reduces the influence of erratic forecasts and invoke a sample-dependent null distribution. The advantages of the new test are confirmed by Monte Carlo experiments and an empirical exercise in which we compare a large number of regression-based forecasts of annual U.S. inflation to a simple random-walk forecast. The random-walk forecast is found to be inferior to regression-based forecasts and, interestingly, the best sample performance is achieved by models that have a Phillips curve structure.

Exchange rates and fundamentals: evidence on the economic value of predictability

A major puzzle in international finance is the well-documented inability of models based on monetary fundamentals to produce better out-of-sample forecasts of the nominal exchange rate than a naive random walk. While this literature has generally employed statistical measures of forecast accuracy, we investigate whether there is any economic value to the predictive power of monetary fundamentals for the exchange rate. We find that, in the context of a simple asset allocation problem, the economic value of exchange rate forecasts from a fundamentals model can be greater than the economic value of random walk forecasts across a range of horizons.

A Test for Superior Predictive Ability

We propose a new test for superior predictive ability. The new test compares favorable to the reality check for data snooping (RC), because the former is more powerful and less sensitive to poor and irrelevant alternatives. The improvements are achieved by two modifications of the RC. We employ a studentized test statistic that reduces the influence of erratic forecasts and invoke a sample dependent null distribution. The advantages of the new test are confirmed by Monte Carlo experiments and in an empirical exercise, where we compare a large number of regression-based forecasts of annual US inflation to a simple random walk forecast. The random walk forecast is found to be inferior to regression-based forecasts and, interestingly, the best sample performance is achieved by models that have a Phillips curve structure.

Biases of Professional Exchange Rate Forecasts: Psychological Explanations and an Experimentally Based Comparison to Novices

The empirical performance of macroeconomic exchange rate models is more than disappointing. This dismal result is also reflected in the forecasting capabilities of professional analysts: all in all, analysts are not in a position to beat naive random walk forecasts. The root for this deficient outcome stems from the fact that professional forecasts are to a large extent influenced by actual changes in exchange rates. A reasonable explanation for this behaviour can be taken from the behavioural finance literature. To test whether this characteristic tends to be general human behaviour in an uncertain environment, we analyse the forecasting behaviour of students experimentally, using a simulated currency series. Our results indicate that a topically oriented trend adjustment behaviour (TOTA) is a general characteristic of human forecasting behaviour. Additionally, we apply a simple model to explain professional and students forecasts.

Bayesian Model Averaging and Exchange Rate Forecasts

Exchange rate forecasting is hard and the seminal result of Meese and Rogoff (1983) that the exchange rate is well approximated by a driftless random walk, at least for prediction purposes, has never really been overturned despite much effort at constructing other forecasting models. However, in several other macro and financial forecasting applications, researchers in recent years have considered methods for forecasting that combine the information in a large number of time series. One method that has been found to be remarkably useful for out-of-sample prediction is simple averaging of the forecasts of different models. This often seems to work better than the forecasts from any one model. Bayesian Model Averaging is a closely related method that has also been found to be useful for out-of-sample prediction. This starts out with many possible models and prior beliefs about the probability that each model is the true one. It then involves computing the posterior probability that each model is the true one, and averages the forecasts from the different models, weighting them by these posterior probabilities. This is effectively a shrinkage methodology, but with shrinkage over models not just over parameters. I apply this Bayesian Model Averaging approach to pseudo-out-of-sample exchange rate forecasting over the last ten years. I find that it compares quite favorably to a driftless random walk forecast. Depending on the currency-horizon pair, the Bayesian Model Averaging forecasts sometimes do quite a bit better than the random walk benchmark (in terms of mean square prediction error), while they never do much worse. The forecasts generated by this model averaging methodology are however very close to (but not identical to) those from the random walk forecast.

The out-of-sample success of term structure models as exchange rate predictors: a step beyond

A large literature suggests that standard exchange rate models cannot outperform a random walk forecast and that the forward rate is not an optimal predictor of the spot rate. However, there is evidence that the term structure of forward premia contains valuable information for forecasting future spot exchange rates and that exchange rate dynamics display nonlinearities. This paper proposes a term-structure forecasting model of exchange rates based on a regime-switching vector equilibrium correction model which is novel in this context. Our model significantly outperforms both a random walk and, to a lesser extent, a linear term-structure vector equilibrium correction model for four major dollar rates across a range of horizons.

Why is it so difficult to beat the random walk forecast of exchange rates?

Recent empirical evidence suggests that the time series behavior of the real exchange rate is well approximated by a nonlinear, exponential smooth transition autoregressive (ESTAR) model. This nonlinearity helps resolve a number of puzzles concerning the persistence and volatility of real exchange rates. In this paper, we explore whether it may also help resolve the well-known difficulties of exchange rate forecasting. We develop a bootstrap test of the random walk hypothesis of the nominal exchange rate, given ESTAR real exchange rate dynamics. We find strong evidence of predictability at horizons of 2 to 3 years, but not at shorter horizons.

Investigating the information content of non-cash-trading index futures using neural networks

Abstract This study investigates the information content of SGX-DT Nikkei 225 and MSCI Taiwan index futures prices during the non-cash-trading (NCT) period. The lead–lag relationship between the futures market during the NCT period and the cash market during its opening period is first investigated by the generalized methods of moments. The obtained leading futures and previous day's cash market closing index are then used as the input variables to predict the opening cash price index by the backpropagation neural network model. Sensitivity analysis is first employed to address and solve the issue of finding the appropriate setup of the topology of the networks. Extensive studies are then performed on the robustness of the constructed network by using different training and testing sample sizes. To demonstrate the effectiveness of our proposed method, the 5-min intraday data of spot and futures index from a 6-month historical record were evaluated using the designed neural network model. Analytic results demonstrate that the proposed neural network model outperforms the neural network model with previous day's closing index as the input variable, the random walk and GARCH model forecasts. It, therefore, indicates that there is valuable information involved in the futures prices during the NCT period that can be used to forecast the opening cash price index. Besides, the neural network model provides better forecasting results than the commonly discussed GARCH model.

Investors' Trade Size and Trading Responses around Earnings Announcements: An Empirical Investigation

Prior research suggests that the earnings expectations of a segment of the market can be described by the seasonal random-walk model. Prior research also provides evidence that less wealthy and less informed investors tend to make smaller trades (small traders) than wealthier and betterinformed investors (large traders). I hypothesize that it is the earnings expectations of small traders that are associated with predictions from the seasonal random-walk model. By directly analyzing the trading activities of small and large traders, this study provides evidence that is largely consistent with the hypotheses. Specifically, small traders' trading response around earnings announcements is increasing in the magnitude of seasonal random-walk forecast errors, even after controlling for absolute analyst forecast errors, contemporaneous price changes, and market-wide trading. Supplementary analysis reveals that this effect is largely confined to firms with relatively impoverished information environments (i.e., smaller firms and firms with little to moderate analyst following).

Why is it so Difficult to Beat the Random Walk Rorecast of Exchange Rates?

We propose an exchange rate model that can explain both the observed volatility and the persistence of real and nominal exchange rate movements and thus in some measure resolves Rogoff's (1996) purchasing power parity puzzle. Our analysis reconciles the well-known difficulties in beating the random walk forecast model with the statistical evidence of nonlinear mean reversion in deviations from fundamentals. Our analysis also provides a compelling rationale for the long-horizon predictability of exchange rates. We find strong empirical support for long-horizon predictability, and we explain why it is difficult to exploit this predictability in real-time forecasts. Our results not only lend support to economists' beliefs that the exchange rate is inherently predictable, but they also help us to understand the reluctance of applied forecasters to abandon chartist methods in favor of models based on economic fundamentals.

Lead indicator models and UK analysts' earnings forecasts

This study examines the predictive ability of models which adjust random walk forecasts of corporate earnings, to incorporate past changes in economic lead indicators. The results suggest that changes in the broad money supply measure M4 contain predictive ability, beyond equivalent changes in other lead indicators or an individual firm's earnings. When forecasts from the broad-money model are compared with forecasts generated by financial analysts a size effect is evident: the superiority of analysts' forecasts is apparent much earlier for large firms than for small firms. This result is consistent with studies suggesting a size-related differential in the collection and dissemination of information by market participants.

A simultaneous, rational expectations model of the Australian dollar/US dollar market

A simultaneous, rational expectations model of the Australian dollar/US dollar foreign exchange market is presented, using information from both spot and futures markets. Tests for unit roots are conducted, and in no case can the hypothesis of a single unit root in either the spot rate or the futures rate be rejected. Cointegration tests also are reported, and following minor respecification of the model, the hypothesis of no cointegration is rejected for all equations. All parameter estimates are of the expected sign, and virtually all parameter estimates are significant. Intra-sample, the model simulates futures and spot exchange rates with per cent RMSEs of 2.7% and 2.6% respectively, while post-sample per cent RMSEs for futures and spot rates are 1.2% and 0.65%. The model significantly outperforms a random walk forecast of the spot rate post-sample, and also significantly surpasses the forecast of a lagged futures rate, this last result being evidence against the efficient markets hypothesis.

Forecasting international conference attendance

There are many tourism modelling/forecasting studies, but most of these are concerned with holiday tourism or total tourism (ie not disaggregated by visit purpose); conference tourism is virtually ignored. In this paper, an econometric model which explains tourism demand for an international conference is developed. The model is specified in logistic form and is estimated using maximum likelihood techniques. Various alternative forms of the model are considered. For the forecasting model selected, good empirical results are obtained in terms of model fit, and, in contrast with previous (non-conference) tourism forecasting findings, the forecasts generated by the econometric model are more accurate than random walk forecasts.

An assessment of the economic value of non‐linear foreign exchange rate forecasts

AbstractRecent research suggests that non‐linear methods cannot improve the point forecasts of high‐frequency exchange rates. These studies have been using standard forecasting criteria such as smallest mean squared error (MSE) and smallest mean absolute error (MAE). It is, however, premature to conclude from this evidence that non‐linear forecasts of high‐frequency financial returns are economically or statistically insignificant. We prove a proposition which implies that the standard forecasting criteria are not necessarily particularly suited for assessment of the economic value of predictions of non‐linear processes where the predicted value and the prediction error may not be independently distributed. Adopting a simple non‐linear forecasting procedure to 15 daily exchange rate series we find that although, when compared to simple random walk forecasts, all the non‐linear forecasts give a higher MSE and MAE, when applied in a simple trading strategy these forecasts result in a higher mean return. It is also shown that the ranking of portfolio payoffs based on forecasts from a random walk, and linear and non‐linear models, is closely related to a non‐parametric test of market timing.