Limited Foresight May Force Cooperation

This paper considers discounted repeated games with boundedly rational players. In each period, player i chooses his current action on the basis of his forecast about the forthcoming n _i action profiles; his assessment of the payoffs he will obtain next depends on his state of mind, which is non-deterministic. A limited forecast equilibrium is such that after every history the limited horizon forecasts formed by the players are correct. The set of all limited forecast equilibria is characterized and constructed. Application to the repeated prisoner's dilemma shows that limited foresight may sometimes induce purely cooperative paths while purely non-cooperative paths cannot arise.     

Monitoring and forecasting annual public deficit every month: the case of France

In this paper we forecast annual budget deficits using monthly information. Using French monthly data on central government revenues and expenditures, the method we propose consists of: (1) estimating monthly ARIMA models for all items of central government revenues and expenditures; (2) inferring the annual ARIMA models from the monthly models; (3) using the inferred annual ARIMA models to perform one-step-ahead forecasts for each item; (4) compounding the annual forecasts of all revenues and expenditures to obtain an annual budget deficit forecast. The major empirical benefit of this technique is that as soon as new monthly data become available, annual deficit forecasts are updated. This allows us to detect in advance possible slippages in central government finances. For years 2002–2004, forecasts obtained following the proposed approach are compared with a benchmark method and with official predictions published by the French government. An evaluation of their relative performance is provided.      

Oil price forecasting under asymmetric loss

Based on the approach developed by Elliott et al. (2005), we found that the loss function of a sample of oil price forecasters is asymmetric in the forecast error. Our findings indicate that the loss oil price forecasters incurred when their forecasts exceeded the price of oil tended to be larger than the loss they incurred when their forecast fell short of the price of oil. Accounting for the asymmetry of the loss function does not necessarily make forecasts look rational.     

Methods for analyzing data from Delphi panels: Some evidence from a forecasting study

Delphi and other methods of using expert opinion to generate forecasts can be a useful tool for planning, impact assessment, and policy analysis. Unfortunately, little is known about the accuracy of forecasts produced using these methods, so their utility is limited at present. Based on the logic of the Delphi method, I suggest that: 1) forecast accuracy should increase across rounds of a Delphi iteration, 2) there is a positive correlation between a panelist's uncertainty about a forecast and his or her shift in forecast from round to round, 3) forecasts weighted by self-reported confidence will be more accurate than unweighted forecasts, and 4) the use of robust estimates of location as summaries of expert opinion yield better forecasts than nonrobust measures. A Delphi experiment provides little support to any of these hypotheses. This finding suggests that traditional assumptions about the proper methods for analyzing a Delphi study may be inappropriate.     

Understanding forecast failure of ESTAR models of real exchange rates

The forecast performance of the empirical ESTAR model of Taylor et al. (2001) is examined for 4 bilateral real exchange rate series over an out-of-sample evaluation period of nearly 12?years. Point as well as density forecasts are constructed, considering forecast horizons of 1 to 22 steps head. The study finds that no forecast gains over a simple AR(1) specification exist at any of the forecast horizons that are considered, regardless of whether point or density forecasts are utilised in the evaluation. Non-parametric methods are used in conjunction with simulation techniques to learn about the models and their forecasts. It is shown graphically that the nonlinearity in the conditional means (or point forecasts) of the ESTAR model decreases as the forecast horizon increases. The non-parametric methods show also that the multiple steps ahead forecast densities are normal looking with no signs of bi-modality, skewness or kurtosis.     

Do we need a global VAR model to forecast inflation and output in South Africa?

This study determines whether the global vector autoregressive (GVAR) approach provides better forecasts of key South African variables than a vector error correction model (VECM) and a Bayesian vector autoregressive (BVAR) model augmented with foreign variables. The article considers both a small GVAR model and a large GVAR model in determining the most appropriate model for forecasting South African variables. We compare the recursive out-of-sample forecasts for South African GDP and inflation from six types of models: a general 33 country (large) GVAR, a customized small GVAR for South Africa, a VECM for South Africa with weakly exogenous foreign variables, a BVAR model, autoregressive (AR) models and random walk models. The results show that the forecast performance of the large GVAR is generally superior to the performance of the customized small GVAR for South Africa. The forecasts of both the GVAR models tend to be better than the forecasts of the augmented VECM, especially at longer forecast horizons. Importantly, however, on average, the BVAR model performs the best when it comes to forecasting output, while the AR(1) model outperforms all the other models in predicting inflation. We also conduct ex ante forecasts from the BVAR and AR(1) models over 2010:Q1–2013:Q4 to highlight their ability to track turning points in output and inflation, respectively.     

Forecasting Housing Approvals in Australia: Do Forecasters Herd?

Price trends in housing markets may reflect herding of market participants. A natural question is whether such herding, to the extent that it occurred, reflects herding in forecasts of professional forecasters. Using more than 6,000 forecasts of housing approvals for Australia, we did not find evidence of forecaster herding. On the contrary, forecasters anti‐herd and, thereby, tend to intentionally scatter their forecasts around the consensus forecast. The extent of anti‐herding seems to vary over time. We also found that more pronounced anti‐herding leads to less accurate forecasts.     

Inefficiency in Earnings Forecasts: Experimental Evidence of Reactions to Positive vs. Negative Information

Prior archival studies of analysts' forecasts have found evidence for systematic underreaction, systematic overreaction, and systematic optimism bias. Easterwood and Nutt (1999) attempt to reconcile the conflicting evidence by testing the robustness of Abarbanell and Bernard's (1992) underreaction results to the nature of the information. Consistent with systematic optimism, forecasts are found to underreact to negative earnings information but overreact to positive information. However, Easterwood and Nutt are unable to distinguish between misreaction caused by incentives unique to analysts with misreaction caused by human decision bias that may be typical of investors. We address this issue by analyzing forecast reactions to positive versus negative information in the controlled experimental setting of Gillette, Stevens, Watts, and Williams (1999). The forecast data reveal systematic underreaction to both positive and negative information, and the underreaction is generally greater for positive information than negative information. This suggests that prior empirical evidence of forecast overreaction to positive information is unlikely to be attributable to human decision bias.     

Toward an Understanding of the Risky Choice Behavior of Professional Financial Analysts

Several studies have reported inefficiencies and/or biases in analysts' ability to incorporate new information into their earnings forecasts. We propose that an important psychological factor associated with optimistic earnings forecasts is the propensity of analysts to engage in risky choice behavior as described by prospect theory. Furthermore, the motivational incentives faced by analysts may exacerbate risky choice behavior during forecast revision, thereby magnifying overestimates of earnings.

Sixty professional financial analysts were asked to issue a first quarter and then an annual EPS forecast of a company. The analysts were randomly assigned to two initial forecast accuracy conditions that indicated their initial forecast earnings was 1) essentially the same as actual earnings, or 2) substantially higher than actual earnings. Analysts were also assigned to one of three motivational incentive conditions indicating the analyst and brokerage firm would 1) have no future contact with the forecast firm, 2) begin to follow the forecast firm, or 3) establish an underwriting relationship with the forecast firm.

The results indicate that analysts who perceived a loss function due to the inaccuracy of prior earnings forecasts tended to choose riskier prospects in subsequent forecast revisions than analysts who perceived their prior earnings forecasts to be accurate. These riskier prospects translate into greater overestimates of earnings. Furthermore, while the average risk attitude of the analysts was optimistic, higher levels of motivational incentives were associated with greater risk-seeking behavior by the analysts who perceive a loss function. It appears that the motivational incentives inherent in brokerage firms can exacerbate the risky choice behavior of financial analysts during forecast revision. These findings support the utility of incorporating both cognitive and motivational factors into the prediction of analyst behavior.     

A Nonlinear Approach for Modeling and Forecasting US Business Cycles

The purpose of this paper is to provide a complete evaluation of four regime-switching models by checking their performance in detecting US business cycle turning points, in replicating US business cycle features and in forecasting US GDP growth rate. Both individual and combined forecasts are considered. Results indicate that while the Markov-switching model succeeded in replicating all the NBER peak and trough dates without an extra-cycle detection, it seems to be outperformed by the Bounce-back model in term of the delay time to a correct alarm. Concerning business cycle features characterization, none of the competing models dominates over all the features. The performance of the Markov-switching and bounce back models in detecting turning points was not translated into an improved business cycle feature characterization since they are outperformed by the Floor and Ceiling model. The forecast performance of the considered models varies across regimes and across forecast horizons. That is, the model performing best in an expansion period is not necessarily the same in a recession period and similarly for the forecast horizons. Finally, combining such individual forecasts generally leads to increased forecast accuracy especially for h=1.     

THE VALUE OF IN‐STREAM WATER TEMPERATURE FORECASTS FOR FISHERIES MANAGEMENT

Water temperature is an important factor affecting aquatic life within the stream environment. Cold water species, such as salmonids, are particularly susceptible to elevated water temperatures. This paper examines the economic value of short‐term water temperature forecasts for salmonid management. Forecasts may have economic value if they allow the water resource manager to make better water allocation decisions. This study considers two applications: water releases for management of Chinook salmon in the Klamath River and leasing water from agriculture for management of steelhead trout in the John Day River. We incorporate biophysical models and water temperature distribution data into a Bayesian framework to simulate changes in fish populations and the corresponding benefit from recreational fishing and opportunity cost of water under different temperature forecast accuracies. Simulation results indicate that use of the forecasts results in increased fish production and that marginal costs decline and net benefits increase as forecast accuracy increases, suggesting that provision and use of such stream temperature forecasts would have value to society. (JEL Q22, Q25, Q28, Q50)     

Excessive Volatility is Also a Feature of Individual Level Forecasts

The excessive volatility of prices in financial markets is one of the most pressing puzzles in social science. It has led many to question economic theory, which attributes beneficial effects to markets in the allocation of risks and the aggregation of information. In exploring its causes, we investigated to what extent excessive volatility can be observed at the individual level. Economists claim that securities prices are forecasts of future outcomes. Here, we report on a simple experiment in which participants were rewarded to make the most accurate possible forecast of a canonical financial time series. We discovered excessive volatility in individual-level forecasts, paralleling the finding at the market level. Assuming that participants updated their beliefs based on reinforcement learning, we show that excess volatility emerged because of a combination of three factors. First, we found that submitted forecasts were noisy perturbations of participants’ revealed beliefs. Second, beliefs were updated using a prediction error based on submitted forecast rather than revealed past beliefs. Third, in updating beliefs, participants maladaptively decreased learning speed with prediction risk. Our results reveal formerly undocumented features in individual-level forecasting that may be critical to understand the inherent instability of financial markets and inform regulatory policy.     

Combination forecasts of tourism demand with machine learning models

The main objective of this study is to analyse whether the combination of regional predictions generated with machine learning (ML) models leads to improved forecast accuracy. With this aim, we construct one set of forecasts by estimating models on the aggregate series, another set by using the same models to forecast the individual series prior to aggregation, and then we compare the accuracy of both approaches. We use three ML techniques: support vector regression, Gaussian process regression and neural network models. We use an autoregressive moving average model as a benchmark. We find that ML methods improve their forecasting performance with respect to the benchmark as forecast horizons increase, suggesting the suitability of these techniques for mid- and long-term forecasting. In spite of the fact that the disaggregated approach yields more accurate predictions, the improvement over the benchmark occurs for shorter forecast horizons with the direct approach.     

Professional forecasts of interest rates and exchange rates: Evidence from the Wall Street Journal’s panel of economists

We analyze economists’ forecasts of interest rates and exchange rates from the Wall Street Journal. We find that a majority of economists produced unbiased forecasts but that none predicted directions of changes more accurately than chance. Most economists’ forecast accuracy is statistically indistinguishable from a random walk model in forecasting the Treasury bill rate, but many are significantly worse in forecasting the Treasury bond rate and the exchange rate. We also find systematic forecast heterogeneity, support for strategic models predicting the industry employing the economist matters, and evidence that economists deviate less from the consensus as they age.     

Outline of forecast theory using generalized cost functions

The cost functions used to form forecasts in practice may be quite different than the squared costs that is often assumed in forecast theory. The impact on evaluation procedures is determined and simple properties for the derivate of the cost function of the errors are found to provide simple tests of optimality. For a very limited class of situations are forecasts based on conditional means optimal, generally, the econometricians needs to provide the whole conditional predicted distribution. Implications for multi-step forecasts and the combination of forecasts are briefly considered.     

Do the OECD 24 month horizon growth forecasts for the G7–countries contain information?

The OECD produces two–year–ahead growth forecasts for the G7–countries since 1987; these forecasts have never been evaluated. A regression is developed that tests for the information content of the forecasts. The idea is that this content is the added value forecasters incorporate in their forecasts. The information content is defined relative to the forecast for the previous year. In the end, the added value contained in the current year forecast is calculated relative to the last observation. The test consists in checking whether the information content reduces the forecasts error.

The study begins with a calculation of the usual accuracy statistics. These indicate an extreme low quality for the forecasts. The regression tests support this conclusion although the forecasts for Japan do possess some information. Alarming for users of forecasts is that there are no obvious alternatives.     

Expectations and optimal monetary policy: A stability problem revisited

A landmark result in the optimal monetary policy design literature is that fundamental-based interest rate rules invariably lead to rational expectations equilibria (REE) that are not stable under adaptive learning. In this paper, we make a novel information assumption that private agents cannot observe aggregate fundamental shocks, and use simple linear forecasting rules for learning. We find that with fundamental-based rules, there exist limited information equilibria that are stable under learning. Moreover, there are multiple equilibria. Learning can be used as a selection tool to identify a unique equilibrium.     

An analysis of a panel of Spanish GDP forecasts

This article undertakes an analysis of the forecasts included in the Panel of Spanish Forecasts in order to highlight the fact that the predictions made and the errors committed by the entities participating on this panel contain information that is particularly useful in analysing the evolution of the Spanish economy. Here, a study is undertaken of the GDP growth forecasts for the Spanish economy for the period 2000 to 2002 made in distinct forecast time horizons. Specifically, it analyses whether the forecasts are optimistic or pessimistic and whether new information concerning the variable being predicted is used efficiently in revising earlier forecasts.     

Was the recent downturn in US real GDP predictable?

This article uses a small set of variables – real GDP, the inflation rate and the short-term interest rate – and a rich set of models – atheoretical (time series) and theoretical (structural), linear and nonlinear, as well as classical and Bayesian models – to consider whether we could have predicted the recent downturn of the US real GDP. Comparing the performance of the models to the benchmark random-walk model by root mean-square errors, the two structural (theoretical) models, especially the nonlinear model, perform well on average across all forecast horizons in our ex post, out-of-sample forecasts, although at specific forecast horizons certain nonlinear atheoretical models perform the best. The nonlinear theoretical model also dominates in our ex ante, out-of-sample forecast of the Great Recession, suggesting that developing forward-looking, microfounded, nonlinear, dynamic stochastic general equilibrium models of the economy may prove crucial in forecasting turning points.     

Multiplicity of Cournot Equilibria and Involuntary Unemployment

In an overlapping generations model with Cournot competition on the goods market it is shown that a continuum of stationary states and perfect foresight trajectories exists with unemployment at arbitrary low wages. Decisive for this is the influence that different forecast functions have on the objective demand curve, even though they are consistent with perfect foresight. With an example it is shown that simple adaptive and constant memory forecast rules generate such unemployment equilibria. The corresponding temporary equilibrium dynamics may display stable unemployment and unstable full employment equilibria.Journal of Economic LiteratureClassification Numbers: D43, D51, E24