Data vs. information: Using clustering techniques to enhance stock returns forecasting

This paper explores the use of clustering models of stocks to improve both (a) the prediction of stock prices and (b) the returns of trading algorithms.We cluster stocks using k-means and several alternative distance metrics, using as features quarterly financial ratios, prices and daily returns. Then, for each cluster, we train ARIMA and LSTM forecasting models to predict the daily price of each stock in the cluster. Finally, we employ the clustering-empowered forecasting models to analyze the returns of different trading algorithms.We obtain three key results: (i) LSTM models outperform ARIMA and benchmark models, obtaining positive investment returns in several scenarios; (ii) forecasting is improved by using the additional information provided by the clustering methods, therefore selecting relevant data is an important preprocessing task in the forecasting process; (iii) using information from the whole sample of stocks deteriorates the forecasting ability of LSTM models.These results have been validated using data of 240 companies of the Russell 3000 index spanning 2017 to 2022, training and testing with different subperiods.     

Affiliated block shareholders and analyst optimism

We investigate several previously under-documented conflicts of interest that may result in analyst optimism by utilizing two unique features of brokerage firms in China, namely, the dominant ownership of large shareholders within the brokerage firms and the mandatory disclosure of brokerage firms’ commission income derived from each mutual fund client. We show that controlling shareholders of an analyst’s brokerage-firm put pressure on the analyst to report more optimistically biased earnings forecasts and recommendations to the stocks they hold larger positions in. We also find that the magnitude of analyst optimism increases with the shareholdings of the mutual funds that contribute commission fees to the analyst’s brokerage firm. These findings remain robust after incorporating a regulation change that reduces conflicts of interest in the brokerage industry and higher dimensional fixed effects, and thus are unlikely to be driven by reverse causality or omitted variable bias.     

How efficient is the market for Australian firms’ earnings information?

We construct a measure of the speed with which forecasts issued by sell-side analysts accurately forecast future annual earnings. Following Marshall, we label this measure earnings information flow timeliness (EIFT). This measure avoids the aggregation problem inherent in price-based measures of information efficiency. We document large variation in EIFT across firm-years, and show that EIFT is positively associated with the extent of analyst following, consistent with increased analyst coverage improving the speed with which earnings-related information is recognised. We also find that EIFT is higher for firm-years classified as ‘bad news’ (i.e., where analysts’ forecasts at the start of the financial period exceed the reported outcome). However, when we separately consider instances where analysts appear to forecast non-GAAP (or ‘street’) earnings rather than GAAP earnings, we find that the greater timeliness of bad news is concentrated among observations where analysts forecast non-GAAP earnings, where unusual items are typically excluded. We conclude that the market for accounting information is more efficient for negative operating outcomes than for negative outcomes reflecting unusual items.     

The effects of analyst forecasts and earnings trends on perceptions of management forecast credibility

We examine whether analyst forecasts influence investors’ perceptions of the credibility of a good news management earnings forecast. We hypothesize that the effect of analyst forecasts will depend on whether the analyst forecast confirms management’s forecast and the extent to which management’s forecast is consistent with the prior earnings trend. Findings indicate that the positive effect of a confirming analyst forecast is greater when the management forecast is trend inconsistent than when it is trend consistent. The negative effect of a disconfirming analyst forecast does not differ based on management forecast trend consistency.     

Economic policy uncertainty and analysts’ forecast characteristics

We examine the influence of economic policy uncertainty (EPU) on the characteristics of analysts’ earnings forecasts over a thirty-year period, spanning a wide variety of political and economic conditions. Motivated by both theory and empirical evidence that suggest a decline in the quality of the information environment for firms as EPU increases, we establish that analysts’ forecast errors increase with EPU, as does the degree of forecast dispersion. Increased error and dispersion persist after controlling for several competing sources of economy-wide uncertainty. Cross sectional analysis exploring heterogeneity in forecast quality across both analyst and firm characteristics establishes that forecast error and dispersion increase with EPU across a broad spectrum of firms and levels of analyst expertise. We control for analysts’ experience overall and the years spent covering a particular industry and firm. Five alternative methods for classifying firms as policy sensitive versus policy neutral provide consistent evidence that analyst forecast errors and dispersion increase with EPU, even for firms not deemed to be particularly sensitive to policy.     

Target price forecasts: The roles of the 52-week high price and recent investor sentiment

This paper reveals that in addition to fundamental factors, the 52-week high price and recent investor sentiment play an important role in analysts’ target price formation. Analysts’ forecasts of short-term earnings and long-term earnings growth are shown to be important explanatory variables for target prices; equally, the 52-week high price and recent investor sentiment are also shown to explain target price levels and especially target price biases. Our analysis additionally reveals that analysts place greater weight on these two non-fundamental factors in settings with greater task complexity and to some extent in those with greater resource constraints. Conversely, on balance, the results suggest that this increased reliance does not translate into an increased impact per unit of each non-fundamental factor on forecast bias. Finally, our results show that target prices are useful in predicting future stock returns beyond earnings forecasts and commonly used risk proxies. However, in an internally consistent fashion, the informativeness of target prices for future returns is significantly reduced when greater weight is placed on either the 52-week high or recent investor sentiment in the target price formation process.     

Analyst earnings forecast precision and local advantage: evidence from American depositary receipt firms

Prior research has documented a finding that local analysts provide more accurate earnings forecasts than nonlocal analysts in many settings. However, little is known about local and nonlocal analysts’ relative earnings forecast abilities for ADR stocks. In this study, we find that the local advantage disappears for ADR stocks and that nonlocal analysts in fact outperform local ones in this case. We investigate the source of this ‘local disadvantage’ and find evidence against hypotheses based on accounting standards and exchange rates. We document that the local advantage decreases with increased investor interest in foreign firms, consistent with our nonlocal investors’ interest hypothesis.     

基于灰色系统理论的常州市物流需求预测

地区性的物流需求预测对于制定宏观经济政策和促进地区经济发展具有重要意义。根据常州市近10年来的物流需求变化,建立了灰色预测的GM（1,1）模型,通过计算对常州市今后10年的物流需求进行了预测,为政府规划地区物流发展提供相关理论参考。     

企业战略差异度会影响分析师盈余预测吗？

随着我国资本市场的发展,证券分析师的作用日益凸显,关于分析师盈余预测的研究一直是各界关注的重点。采用我国沪深两市A股上市公司2011—2015年分析师盈余预测数据,检验了企业战略差异度对分析师盈余预测精度和分析师跟踪偏好的影响。实证结果表明：（1）企业战略差异度越大,分析师盈余预测准确度越低;（2）企业战略差异度越大,分析师盈余预测分歧度越大;（3）企业战略差异度越大,分析师跟踪人数越少。进一步研究还发现：国有产权会削弱战略差异度对分析师盈余预测准确度和跟踪人数的不利影响。     

基于等维新息灰色马尔可夫模型的河南旅游总收入预测

以河南省旅游总收入为对象进行研究,选取2005—2018年的旅游总收入数据作为研究样本,在GM(1,1)模型的基础上,用马尔可夫理论和等维新息的思想加以修正,建立了等维新息灰色马尔可夫模型。结果表明,等维新息灰色马尔可夫模型比传统的GM（1,1）模型和灰色马尔可夫模型的预测精度要高,达到了94.735%,可用来进行旅游总收入的预测,从而为有关部门制定政策提供参考。