本文選題：機構(gòu)投資者調(diào)研 + 管理層盈余預(yù)測�。� 參考：《管理科學(xué)》2017年01期

【摘要】：管理層盈余預(yù)測作為改善公司信息環(huán)境的重要手段之一,受到研究者們的廣泛關(guān)注,近年來相關(guān)研究越來越強調(diào)具有信息優(yōu)勢的管理層會對業(yè)績預(yù)告的披露時間和方式進行戰(zhàn)略性選擇,以實現(xiàn)自身利益最大化的目的。因此,對如何抑制管理層利用信息優(yōu)勢謀取私利、提高盈余預(yù)測精確度這一問題進行研究具有重要意義。從信息不對稱視角入手,以2013年至2015年深圳交易所披露管理層盈余預(yù)測的上市公司為研究樣本,采用順序Logit模型等多元回歸分析方法,實證檢驗機構(gòu)投資者調(diào)研與管理層盈余預(yù)測方式的關(guān)系,并進一步考察盈余預(yù)測信息性質(zhì)和產(chǎn)權(quán)性質(zhì)對二者關(guān)系的影響。以每個上市公司一年內(nèi)接待的機構(gòu)投資者調(diào)研次數(shù)或一年內(nèi)接待的機構(gòu)投資者總數(shù)的自然對數(shù)測量機構(gòu)投資者調(diào)研頻率。根據(jù)管理層盈余預(yù)測精確度將管理層預(yù)測分為4類,即當管理層盈余預(yù)測為點估計時賦值為4,為閉區(qū)間估計時賦值為3,為開區(qū)間估計時賦值為2,為定性預(yù)測時賦值為1。研究結(jié)果表明,總體而言機構(gòu)投資者調(diào)研越頻繁的公司,其管理層越傾向于采用模糊的方式預(yù)測盈余,并且這種影響在盈余預(yù)測信息為壞消息時更為明顯,以上情形主要存在于非國有企業(yè)中。進一步檢驗發(fā)現(xiàn),相對于買方機構(gòu)而言,賣方機構(gòu)的調(diào)研行為更有可能導(dǎo)致管理層采用模糊方式進行盈余預(yù)測,公募基金和私募基金的調(diào)研行為對預(yù)測方式的影響并無顯著差異。上述研究結(jié)果在進行工具變量兩階段最小二乘回歸、Heckman兩階段檢驗等一系列穩(wěn)健性檢驗后依然成立。機構(gòu)投資者作為重要的市場參與者,其行為一直受到廣泛關(guān)注,與已有認為機構(gòu)投資者與管理層直接溝通能夠有助于降低信息不對稱程度的研究不同,研究結(jié)果表明機構(gòu)投資者調(diào)研給公司帶來了業(yè)績壓力,從而使其更傾向于選擇模糊的方式進行盈余預(yù)測,拓展了機構(gòu)投資者行為對資本市場信息環(huán)境影響的相關(guān)研究,同時為中國投資者更有效地理解和利用管理層業(yè)績預(yù)告信息進行決策、監(jiān)管機構(gòu)進一步完善上市公司信息披露規(guī)定、保護中小投資者權(quán)益提供了經(jīng)驗借鑒。
[Abstract]:As one of the important means to improve the information environment of the company, the management earnings prediction has been widely concerned by researchers. In recent years, more and more researches have emphasized that managers with information advantages will make strategic choices on the time and mode of disclosure of performance forecasts, in order to achieve the goal of maximizing their own interests. Therefore, it is of great significance to study how to restrain the management from making use of information advantages for private gain and to improve the accuracy of earnings prediction. From the perspective of information asymmetry, this paper takes listed companies which disclose management earnings forecast in Shenzhen Stock Exchange from 2013 to 2015 as research samples, and adopts multiple regression analysis, such as sequential logit model, etc. The empirical study examines the relationship between institutional investor research and management earnings forecasting, and further examines the impact of the nature of earnings forecasting information and property rights on the relationship between the two. The natural logarithm of the total number of institutional investors received by each listed company within one year or the number of institutional investors received within one year measures the frequency of institutional investor research. According to the accuracy of management earnings prediction, the management forecast is divided into four categories, namely, when the management earnings forecast is a point estimate, the value is assigned to 4, the value to the closed interval estimation is 3, the value to the open interval estimate is 2, and the value to the qualitative prediction is 1. The results show that, in general, the more frequent the institutional investors survey, the more the managers tend to forecast earnings in a fuzzy way, and this effect is more obvious when the earnings forecast information is bad news. The above situation mainly exists in non-state-owned enterprises. Further tests show that the research behavior of the seller's organization is more likely to lead the management to make the earnings prediction in a fuzzy way than the buyer's organization. There is no significant difference between the research behavior of public funds and private funds. The above results still hold true after a series of robust tests such as two-stage least square regression and Heckman two-stage test of tool variables. As an important market participant, the behavior of institutional investors has been widely concerned, unlike the existing research that direct communication between institutional investors and management can help reduce the degree of information asymmetry. The results show that institutional investor research brings performance pressure to the company, which makes it more inclined to choose a fuzzy way to predict earnings, and extends the relevant research on the impact of institutional investor behavior on the information environment of capital market. At the same time, it provides experience and reference for Chinese investors to understand and use management performance forecast information more effectively, regulators to further improve the information disclosure regulations of listed companies, and to protect the rights and interests of small and medium-sized investors.
【作者單位】： 北京交通大學(xué)經(jīng)濟管理學(xué)院;
【基金】：國家自然科學(xué)基金(71272055,71572009)~~
【分類號】：F275

【相似文獻】

相關(guān)期刊論文 前1條

1 趙良;都市影像[J];新西部;2001年06期

相關(guān)會議論文 前1條

1 李劍嵐;;淺談產(chǎn)能精準化預(yù)測方式[A];上海市煙草系統(tǒng)2012年度優(yōu)秀學(xué)術(shù)論文集（經(jīng)濟管理類）[C];2011年

，

本文編號：2068593