预测中国宏观经济变量:专家与模型的组合预测

本文使用组合预测方法,探究以“朗润预测”为代表的专家预测以及计量模型对于中国宏观经济变量的预测效果,并研究对不同预测进行组合预测是否有助于改进预测效果。本文发现,对我国CPI和GDP的增长率,专家预测效果总体上优于模型预测。从原因看,一方面,专家在预测时已经考虑了计量模型的预测信息;另一方面,在经济出现“拐点”的时期,专家通过对实际经济环境和政策的把握,得出更准确的经济预测。组合预测有助于提升预测精度,对专家预测进行组合得到的预测效果优于大多数的专家预测,“模型—专家”组合预测的效果也优于所有的模型和大部分专家预测。     

Flu Fallout: Information Production Constraints and Corporate Disclosure

Using influenza epidemic data, we examine how constraints on corporate information production affect disclosure policies. We find that firms in areas with higher flu activity are less likely to issue short-run earnings forecasts and more likely to issue long-run earnings forecasts. These results are more pronounced when the information production process is more complex, when managers face a greater reputational loss for issuing low-quality short-run forecasts, and when firms’ costs of switching the forecast horizon are lower. Further analysis implies that the effect of flu activity on these forecast issuance decisions is not driven by firm performance or information uncertainty. Our results suggest that managers do not simply avoid issuing forecasts in response to information production constraints. Instead, they shift the forecast horizon from short-run to long-run, appearing to balance the costs of issuing low-quality forecasts with those of not issuing forecasts at all.     

Jackknifing stock return predictions

We show that the general bias-reducing technique of jackknifing can be successfully applied to stock return predictability regressions. Compared to standard OLS estimation, the jackknifing procedure delivers virtually unbiased estimates with mean squared errors that generally dominate those of the OLS estimates. The jackknifing method is very general, as well as simple to implement, and can be applied to models with multiple predictors and overlapping observations. Unlike most previous work on inference in predictive regressions, no specific assumptions regarding the data generating process for the predictors are required. A set of Monte Carlo experiments show that the method works well in finite samples and the empirical section finds that out-of-sample forecasts based on the jackknife estimates tend to outperform those based on the plain OLS estimates. The improved forecast ability also translates into economically relevant welfare gains for an investor who uses the predictive regression, with jackknife estimates, to time the market.     

STAR and ANN models: forecasting performance on the Spanish “Ibex-35” stock index

This paper studies whether it is possible to exploit the nonlinear behaviour of daily returns on the Spanish Ibex-35 stock index returns to improve forecasts over short and long horizons. In this sense, we examine the out-of-sample forecast performance of smooth transition autoregression (STAR) models and artificial neural networks (ANNs). We use one-step (obtained by using recursive and nonrecursive regressions) and multi-step-ahead forecasting methods. The forecasts are evaluated with statistical and economic criteria. In terms of statistical criteria, we compared the out-of-sample forecasts using goodness of forecast measures and various testing approaches. The results indicate that ANNs consistently surpass the random walk model and, although the evidence for this is weaker, provide better forecasts than the linear AR model and the STAR models for some forecast horizons and forecasting methods. In terms of the economic criteria, we assess the relative forecast performance in a simple trading strategy including the impact of transaction costs on trading strategy profits. The results indicate a better fit for ANN models, in terms of the mean net return and Sharpe risk-adjusted ratio, by using one-step-ahead forecasts. These results show there is a good chance of obtaining a more accurate fit and forecast of the daily stock index returns by using one-step-ahead predictors and nonlinear models, but that these are inherently complex and present a difficult economic interpretation.     

Can the information content of share repurchases improve the accuracy of equity premium predictions?

We adjust the dividend–price ratio for share repurchases and investigate whether predictive power can be improved when constructing forecasts of the UK and French equity premia. Regulations in the two largest European stock markets allow us to employ actual repurchase data in our predictive regressions. Hence, we are able to overcome problems associated with markets characterised by less stringent disclosure requirements, where investors might have to rely on proxies for measuring repurchase activity. We find that predictability does not improve either in a statistical or in an economically significant sense once actual share repurchases are considered. Furthermore, we employ a proxy measure of repurchases which can be easily constructed in international markets and demonstrate that its predictive content is not in line with that of the actual repurchase data.     

Out-of-Sample Equity Premium Predictability in South Africa: Evidence from a Large Number of Predictors

This article uses a predictive regression framework to examine the out-of-sample predictability of South Africa’s equity premium, using a host of financial and macroeconomic variables. We employ various methods of forecast combination, bootstrap aggregation (bagging), diffusion index (principal component), and Bayesian regressions to allow for a simultaneous role of the variables under consideration, besides individual predictive regressions. We assess both the statistical and economic significance of the individual predictive regressions, combination methods, bagging, principal components, and Bayesian regressions. Our results show that forecast combination methods and principal component regressions improve the predictability of the equity premium relative to the benchmark autoregressive model of order one (AR[1]). However, the Bayesian predictive regressions are found to be the standout performers with the models outperforming the individual regressions, forecast combination methods, bagging and principal component regressions, both in terms of statistical (forecasting) and economic (utility) gains.     

The Effect of Managerial Litigation Risk on Earnings Warnings: Evidence from a Natural Experiment

We examine the causal effect of managerial litigation risk on managers’ disclosure of earnings warnings in the face of large earnings shortfalls. Exploring the staggered adoption of universal demand (UD) laws as an exogenous decrease in litigation risk, we find that the adoption leads to a decrease in managers’ issuance of earnings warnings, especially among firms facing a higher litigation risk prior to the adoption. In contrast, we find no change in managers’ tendency to alert investors of impending large positive earnings surprises. Collectively, our results provide causal evidence that higher litigation risk incentivizes managers to issue more earnings warnings. Our results differ from Bourveau et al.’s finding of an increase in the frequency of management earnings forecasts after the adoption of UD laws. We reconcile our findings with theirs by demonstrating that the effect of adopting UD laws on management earnings forecasts depends critically on forecast horizon: The adoption increases long-horizon forecasts, but decreases short-horizon forecasts.     

Determinants of analysts’ revenue forecast accuracy

This study investigates financial analysts’ revenue forecasts and identifies determinants of the forecasts’ accuracy. We find that revenue forecast accuracy is determined by forecast and analyst characteristics similar to those of earnings forecast accuracy—namely, forecast horizon, days elapsed since the last forecast, analysts’ forecasting experience, forecast frequency, forecast portfolio, reputation, earnings forecast issuance, forecast boldness, and analysts’ prior performance in forecasting revenues and earnings. We develop a model that predicts the usefulness of revenue forecasts. Thereby, our study helps to ex ante identify more accurate revenue forecasts. Furthermore, we find that analysts concern themselves with their revenue forecasting performance. Analysts with poor revenue forecasting performance are more likely to stop forecasting revenues than analysts with better performance. Their decision is reasonable because revenue forecast accuracy affects analysts’ career prospects in terms of being promoted or terminated. Our study helps investors and academic researchers to understand determinants of revenue forecasts. This understanding is also beneficial for evaluating earnings forecasts because revenue forecasts reveal whether changes in earnings forecasts are due to anticipated changes in revenues or expenses.     

Analyst revenue forecast reporting and the quality of revenues and expenses

We decompose earnings quality into revenue and expense quality and examine their associations with analyst propensity to supplement their earnings forecasts with revenue forecasts. Analysts report more revenue forecasts to I/B/E/S when expense quality is low to compensate for the low accuracy of their earnings estimates, which has a positive association with expense quality. Expense quality is unassociated with revenue forecast accuracy, thus revenue forecasts become increasingly useful for valuing firms when expense quality is low. Analysts report fewer revenue forecasts when revenue quality is low because both earnings and revenue forecast accuracy decline as revenue quality deteriorates. To control for endogeneity, we use firm‐fixed effects to control for unobserved time‐invariant heterogeneity across firms, instrumental variables regressions and regression in changes.     

How Disaggregation Enhances the Credibility of Management Earnings Forecasts

An important problem facing managers is how to enhance the credibility, or believability, of their earnings forecasts. In this paper, we experimentally test whether a characteristic of a management earnings forecast—namely, whether it is disaggregated—can affect its credibility. We also test whether disaggregation moderates the relation between managerial incentives and forecast credibility. Disaggregated forecasts include an earnings forecast as well as forecasts of other key line items comprising that earnings forecast. Our results indicate that disaggregated forecasts are judged to be more credible than aggregated ones and that disaggregation works to counteract the effect of high incentives. We also develop and test an original model that explains how disaggregation positively impacts three factors that, in turn, influence forecast credibility: perceived precision of management's beliefs, perceived clarity of the forecast, and perceived financial reporting quality. We show that forecast disaggregation works to remedy incentive problems only via its effect on perceived financial reporting quality. Overall, our study adds to our understanding of how managers can credibly communicate their expectations about the future to market participants.     

Option augmented density forecasts of market returns with monotone pricing kernel

Basic financial theory indicates that the ratio of the conditional density of the future value of a market index and the corresponding risk neutral density should be monotone, but a sizeable empirical literature finds otherwise. We therefore consider an option augmented density forecast of the market return obtained by transforming a baseline density forecast estimated from past excess returns so as to monotonize its ratio with a risk neutral density estimated from current option prices. To evaluate our procedure, we compare baseline and option augmented monthly density forecasts for the S&P 500 index over the period 1997–2013. We find that monotonizing the pricing kernel leads to a modest improvement in the calibration of density forecasts. Supplementary results supportive of this finding are given for market indices in France, Germany, Hong Kong, Japan and the UK.     

An examination of the effects of management earnings forecast form and explanations on financial analyst forecast revisions

Using a sample of 978 quarterly management earnings-per-share forecasts made during the period 1993 to 1999, we document that financial analyst revisions to management earnings forecasts are a function of management forecast form. More precise forecasts (measured three different ways) lead to greater revision of financial analyst consensus EPS forecasts for a given level of unexpected earnings as predicted by Kim and Verrecchia (1991) and Bayesian adjustment models. Also, consistent with our arguments, maximum forecasts are interpreted as bad news by analysts. Our results, while consistent with theory, are inconsistent with recent experimental studies which do not reject the null hypothesis of no effect of management earnings forecast form on the association between unexpected earnings and financial analyst forecast revisions. We also re-examine Baginski, Hassell, and Kimbrough's (2004) finding that attributions used to explain management forecasts affect the reaction to the forecast using analyst data. Consistent with their findings using stock prices, the attribution presence (especially external attributions) increases financial analyst revisions pursuant to management forecasts.     

Analysts' Rationality and Forecast Bias: Evidence from Sales Forecasts

When optimistic forecasts can improve access to management, rational analysts have incentives to issue optimistically-biased forecasts (Lim, 2001). This paper proposes that the extent of this optimistic forecast bias will depend on the forecast's importance to management. If management attaches less importance to a forecasted measure, analysts should decrease their forecast bias because the expected benefits of issuing optimistic forecasts are less. We examine analysts' earnings and sales forecasts, and predict that analysts' optimistic bias will be greater for earnings than for sales. Results are consistent with our predictions and contribute to the evidence that analysts' forecast bias is rational and intentional.     

Accounting based valuation: a simultaneous equations model for forecasting earnings to proxy for ‘other information’

This paper develops and tests a simultaneous equations model (SEM) for extending accounting based valuation models used in empirical studies. Rather than using analysts’ forecasts, we derive forecasts of operating income from the SEM to calculate the ‘other information’ variable in the Ohlson (Contemp Account Res 11:661–687, 1995) model. The SEM forecasts are based on observable data contained in the firms’ reporting, like order backlog, and other publicly available information. The SEM produces more accurate out-of-sample forecasts of operating income compared to simple benchmark models particularly in years around economic changes and instability, like the years 2001 and 2009. Integrating the SEM forecast as ‘other information’ in market value regressions significantly increases the explanatory power compared to simpler versions without or with single information proxies for ‘other information’. Finally, we find that the SEM forecast is able to explain a major portion of the information advantage of analysts relevant for explaining market values.     

Analyst reputation and management earnings forecasts

Prior studies show that analysts with high reputation are influential in the market. This paper examines whether managers consider analyst reputation in shaping their voluntary disclosure strategy. Using Institutional Investor magazine’s All-American (AA) rankings as a proxy for analyst reputation, we find that the coverage of AA analysts is positively associated with the likelihood of quarterly management earnings forecasts (MEFs). We also find that AA analysts’ forecast optimism is more positively associated with the likelihood of MEFs than non-AA analysts’ forecast optimism when the firm is covered by AA analysts. Analyses based on AA analyst coverage changes and AA status changes confirm the relation between analyst reputation and MEFs. We further find that analyst reputation influences other MEF properties, such as forecast news, bias, and revisions, and that our results are robust to alternative measures of analyst reputation. Further analyses show that market reactions at quarterly earnings announcements are more positive (negative) when firms meet/beat (miss) AA analysts’ forecasts than when firms meet/beat (miss) non-AA analysts’ forecasts. Collectively, our findings suggest that managers strategically provide voluntary forecasts by taking into account the reputation of individual analysts following their firms.     

Framing Effects in Stock Market Forecasts: The Difference Between Asking for Prices and Asking for Returns

Studies analyzing return expectations of financial market participantslike fund managers, CFOs or individual investors are highlyinfluential in academia and practice. We argue and show thatthe results in these surveys above are easily influenced bythe elicitation mode of return expectations. Surveys that askfor future stock price levels are more likely to produce meanreverting expectations than surveys that directly ask for futurereturns. Furthermore, we conduct a questionnaire study thatexplicitly analyzes whether the specific elicitation mode affectsreturn expectations in the above direction. In our study, subjectswere asked to state mean forecasts for seven time series. Usinga between subject design, one half of the subjects was askedto state future price levels, the other group was directly askedfor returns. We observe a highly significant framing effect.For upward sloping time series, the return forecasts statedby investors in the return forecast mode are significantly higherthan those derived for investors in the price forecast mode.For downward sloping time series, the return forecasts givenby investors in the return forecast mode are significantly lowerthan those derived for investors in the price forecast mode.We argue that this finding is consistent with behavioral theoriesof investor expectation formation based on the representativenessheuristic.     

Are more able managers good future tellers? Learning from Japan

Baik et al. (2011) find that high-ability managers in the U.S. are more likely to issue accurate management earnings forecasts. Focusing on Japan, where management earnings forecasts are effectively mandated, we extend the literature by exploring (1) whether the relationship between managerial ability and forecast accuracy is unique to the U.S. disclosure system, where management forecasts are voluntary, and (2) how high-ability managers increase their forecast accuracy. We find that managerial ability is negatively associated with forecast errors based on initial forecasts, suggesting that high-ability managers are more likely to issue accurate forecasts at the beginning of the fiscal year. We then show that high-ability managers are less likely to revise their initial earnings forecasts and less likely to use earnings management to improve the accuracy of their earnings forecasts. Our findings show that, while high-ability managers are more likely to issue accurate initial management forecasts, low-ability managers are more likely to revise their forecasts and conduct earnings management to reduce their forecast errors.     

The Credibility of Earnings Forecasts in IPO Prospectuses and Underpricing

This paper provides empirical evidence of the impact of the voluntary disclosure of management earnings forecasts in IPO prospectuses and of the credibility of these forecasts, as perceived by investors at the time of the IPO. We measure forecast credibility ex ante with two approaches: (i) a vector of determinants of credibility that are observable by market participants at the time of the issue and (ii) the predicted value of the forecast error based on some of these determinants. Controlling for the firm's decision on whether or not to issue a forecast, we find that the issue of a forecast reduces underpricing. We find that the quality of the firm's governance and of the auditor and underwriter associated with the issue seems to act as a substitute to the disclosure of an earnings forecast in the prospectus, so that they significantly decrease the level of underpricing only for non‐forecasters. However, despite our various approaches to measure ex ante credibility, we find no association between the pricing of the issue and perceived forecast credibility at the time of the IPO.     

Analyst Pessimism and Forecast Timing

In this study, we show that on average relatively pessimistic analysts tend to reveal their earnings forecasts later than other analysts. Further, we find this forecast timing effect explains a substantial proportion of the well‐known decrease in consensus analyst forecast optimism over the forecast period prior to earnings announcements, which helps explain why analysts’ longer term earnings forecasts are more optimistically biased than their shorter term forecasts. We extend the theory of analyst self‐selection regarding their coverage decisions to argue that analysts with a relatively pessimistic view–compared to other analysts–are more reluctant to issue their earnings forecasts, with the result that they tend to defer revealing their earnings forecasts until later in the forecasting period than other analysts.     

Forecast accuracy in initial public offerings and the impact of external constraints relative to managerial choice: a research note

We examine the accuracy of earnings forecasts published in prospectuses, under conditions where external influences of state regulation and economic condition act as constraints to mitigate management optimism. Our results indicate that regulation has no significant impact, but economic condition and management optimism are significantly associated with forecast accuracy. We conclude that a study of forecast accuracy over time must take into account external influences and that attempting to use regulation to improve forecast accuracy is not an effective strategy when there are strongly adverse economic conditions and when promoters or managers are optimistic in forecasting.