This dissertation examines the general research issue of whether the components of earnings are informative and specifically 1) how analysts consider earnings components when predicting future earnings and 2) whether the information content in, and analysts’ use of, earnings components have changed through time. Although earnings components have predictive value for future earnings based on each component’s persistence, extant research provides only a limited understanding of whether and how analysts consider this when forecasting. Using an integrated income statement and balance sheet framework to estimate the persistence of earnings components, I first establish that disaggregation based on the earnings components framework in this study is helpful to predict future earnings and helps explains contemporaneous returns. I then find evidence suggesting that although analysts consider the persistence of various earnings components, they do not fully integrate this information into their forecasts. Interestingly, analysts appear to be selective in their incorporation of the information in earnings components, seeming to ignore information from components indicating lower persistence, which results in higher forecast errors. Conversely, when a firm’s income is concentrated in high persistence items, analysts appear to incorporate the information into their forecasts, reducing their forecast errors. I also report that the usefulness of components relative to aggregate earnings has dramatically and continuously increased over the past several decades, and contemporaneous returns appear to be much better explained by earnings components than aggregate earnings (than historically). Finally, the relation between analyst forecast errors and the differential persistence of earnings components has also declined over time, indicating that analysts appear to recognize the increasing importance of earnings components through time. / text

Ao longo da última década, tem aumentado o reconhecimento de medidas de gerenciamento de valor. Dentre essas, uma que tem recebido grande atenção tanto no meio acadêmico quanto nas empresas em geral é o valor econômico agregado (EVA®). Muito se tem discutido sobre essa medida, sendo que seus defensores afirmam que ela é uma melhor medida de desempenho do que as medidas contábeis tradicionais. Nessa perspectiva, uma série de pesquisas tem sido realizada, verificando a relação entre o EVA® e o retorno de ações, onde os resultados alternam-se entre uma relação superior dessa medida e o retorno de ações em comparação com as medidas contábeis tradicionais e uma fraca relação ou a ausência de relação entre essas variáveis. Em diferente abordagem, Machuga, Pfeiffer Jr. e Verma (2002) realizaram um estudo no mercado norte americano para verificar a capacidade do EVA® na predição de lucros futuros. Replicando a metodologia desse estudo, esta pesquisa teve como objetivo verificar empiricamente se o EVA® fornece informação incremental para predição de lucros futuros das empresas de capital aberto do Brasil. Na metodologia, foram aplicados modelos de regressão linear múltipla no período de 1998 a 2006 para testar a proposição de que o EVA® fornece informação incrementalmente útil para predizer lucros de um ano adiante das empresas de capital aberto do Brasil. Foram aplicadas regressões anuais (crosssection) e verificou-se a significância estatística dos coeficientes médios. Com os resultados obtidos, não se pode comprovar a utilidade incremental do EVA® na predição de lucros futuros. Na seqüência, realizou-se um teste do valor incremental da inclusão da informação EVA® no modelo de predição, sendo que foram aplicadas novas regressões sem as variáveis EVA® e apurados os novos coeficientes médios; em seguida, foram efetuadas duas previsões de lucros, uma utilizando os valores médios com e outra sem o EVA® . Por meio da comparação desses valores previstos com os reais dos lucros e apurando suas respectivas diferenças, obteve-se os erros médios de previsão. Foi observado que os erros médios de previsão apresentaram-se elevados em função da alta dispersão das variáveis da pesquisa, também foi encontrado que os erros médios de previsão foram menores quando houve a inclusão da informação do EVA® , indicando a utilidade incremental dessa medida na predição de lucros futuros, entretanto esses resultados devem ser interpretados como indicativos e não como conclusivos, já que os coeficientes das variáveis, em sua maioria, não se demonstraram estatisticamente significantes. / There has been increased recognition over the last decade of the measures of management of value. Among these, one that has received the great attention either on the academic field or in the companies in general is the Economic Value Added (EVA®). A lot has been argued about this measure, its defenders affirm that it is one measure of performance better than the traditional accounting measures. In these perspective, a lot of researches have been done, verifying the relation between the EVA® and the stock returns, where the results change between one relation superior of these measures and the stock returns in comparison with the usual accounting measures and a weak relationship or absence of relation between these variables. In a different approach, Machuga, Pfeiffer Jr. and Verma (2002) realize a study on the North America market to verify the ability of EVA® in the prediction of future earnings. Applying the methodology of this study, this research had as goal to verify empirically if the EVA® supplies incremental information to predict future earnings of the Brazilian public companies. After, in the methodology, some multiple linear regression models were applied on the period of 1998 to 2006 to test the proposition that EVA® supplies information incrementally useful to predict one-year-ahead earnings of the Brazilian public companies. The annual cross-section regressions were applied and verified the statistic significance of the average coefficients. With the gotten results, one cannot confirm the incremental utility of EVA® in the future earnings prediction. In the sequence, a test of the incremental value of the inclusion of the information EVA® on the model of prediction was realized, it being that news regressions were applied without the variables EVA® and gotten the new average coefficients, after that, two predictions of earnings was effected, one using the mean values with and the other without the EVA® information. By the comparison of the predicted values with the actual earnings and checking its respective differences, one got the average forecast errors. It was observed that the average forecast errors had been presented high in function of the high dispersion of the variables of the research. It was founded too that the average forecast errors were lower when was included the information of EVA®, indicating the incremental utility of this measure on the prediction of future earnings, however, these results must be interpreted as indicative and not as conclusive, since the coefficients of the variables, in its majority, did not show statistically significant.

Ao longo da última década, tem aumentado o reconhecimento de medidas de gerenciamento de valor. Dentre essas, uma que tem recebido grande atenção tanto no meio acadêmico quanto nas empresas em geral é o valor econômico agregado (EVA®). Muito se tem discutido sobre essa medida, sendo que seus defensores afirmam que ela é uma melhor medida de desempenho do que as medidas contábeis tradicionais. Nessa perspectiva, uma série de pesquisas tem sido realizada, verificando a relação entre o EVA® e o retorno de ações, onde os resultados alternam-se entre uma relação superior dessa medida e o retorno de ações em comparação com as medidas contábeis tradicionais e uma fraca relação ou a ausência de relação entre essas variáveis. Em diferente abordagem, Machuga, Pfeiffer Jr. e Verma (2002) realizaram um estudo no mercado norte americano para verificar a capacidade do EVA® na predição de lucros futuros. Replicando a metodologia desse estudo, esta pesquisa teve como objetivo verificar empiricamente se o EVA® fornece informação incremental para predição de lucros futuros das empresas de capital aberto do Brasil. Na metodologia, foram aplicados modelos de regressão linear múltipla no período de 1998 a 2006 para testar a proposição de que o EVA® fornece informação incrementalmente útil para predizer lucros de um ano adiante das empresas de capital aberto do Brasil. Foram aplicadas regressões anuais (crosssection) e verificou-se a significância estatística dos coeficientes médios. Com os resultados obtidos, não se pode comprovar a utilidade incremental do EVA® na predição de lucros futuros. Na seqüência, realizou-se um teste do valor incremental da inclusão da informação EVA® no modelo de predição, sendo que foram aplicadas novas regressões sem as variáveis EVA® e apurados os novos coeficientes médios; em seguida, foram efetuadas duas previsões de lucros, uma utilizando os valores médios com e outra sem o EVA® . Por meio da comparação desses valores previstos com os reais dos lucros e apurando suas respectivas diferenças, obteve-se os erros médios de previsão. Foi observado que os erros médios de previsão apresentaram-se elevados em função da alta dispersão das variáveis da pesquisa, também foi encontrado que os erros médios de previsão foram menores quando houve a inclusão da informação do EVA® , indicando a utilidade incremental dessa medida na predição de lucros futuros, entretanto esses resultados devem ser interpretados como indicativos e não como conclusivos, já que os coeficientes das variáveis, em sua maioria, não se demonstraram estatisticamente significantes. / There has been increased recognition over the last decade of the measures of management of value. Among these, one that has received the great attention either on the academic field or in the companies in general is the Economic Value Added (EVA®). A lot has been argued about this measure, its defenders affirm that it is one measure of performance better than the traditional accounting measures. In these perspective, a lot of researches have been done, verifying the relation between the EVA® and the stock returns, where the results change between one relation superior of these measures and the stock returns in comparison with the usual accounting measures and a weak relationship or absence of relation between these variables. In a different approach, Machuga, Pfeiffer Jr. and Verma (2002) realize a study on the North America market to verify the ability of EVA® in the prediction of future earnings. Applying the methodology of this study, this research had as goal to verify empirically if the EVA® supplies incremental information to predict future earnings of the Brazilian public companies. After, in the methodology, some multiple linear regression models were applied on the period of 1998 to 2006 to test the proposition that EVA® supplies information incrementally useful to predict one-year-ahead earnings of the Brazilian public companies. The annual cross-section regressions were applied and verified the statistic significance of the average coefficients. With the gotten results, one cannot confirm the incremental utility of EVA® in the future earnings prediction. In the sequence, a test of the incremental value of the inclusion of the information EVA® on the model of prediction was realized, it being that news regressions were applied without the variables EVA® and gotten the new average coefficients, after that, two predictions of earnings was effected, one using the mean values with and the other without the EVA® information. By the comparison of the predicted values with the actual earnings and checking its respective differences, one got the average forecast errors. It was observed that the average forecast errors had been presented high in function of the high dispersion of the variables of the research. It was founded too that the average forecast errors were lower when was included the information of EVA®, indicating the incremental utility of this measure on the prediction of future earnings, however, these results must be interpreted as indicative and not as conclusive, since the coefficients of the variables, in its majority, did not show statistically significant.

På vilket sätt är redovisningsmått som publiceras i företagens årsredovisningar relevanta för att bestämma aktievärden? Kan redovisningsmått användas för att utforma lönsamma placeringsstrategier i aktier? Frågor som dessa är av intresse för såväl akademiker som professionellt verksamma placerare.  I denna avhandling utreds huruvida publicerade redovisningsmått är värderelevanta, i betydelsen att de kan användas för prognoser av företagens framtida räntabilitet på eget kapital. Statistiska modeller för prognos av räntabilitet på eget kapital med hjälp av redovisningsbaserade nyckeltal har estimerats och utvärderats. Det empiriska datamaterialet har utgjorts av årsredovisningar för svenska rörelsedrivande företag under perioden 1970-1985.  Vidare studeras om placeringsstrategier baserade på prognoser av framtida räntabilitet på eget kapital kan ge en aktieavkastning utöver vad som motiveras av placeringens risk. I denna del av studien prövas huruvida den svenska aktiemarknaden är informationseffektiv med avseende på offentligt tillgänglig årsredovisningsinformation. Placeringstrategier har utvärderats på aktier som fanns noterade på Stockholms fondbörs under perioden 1983-1994.  Den empiriska kartläggningen indikerar att redovisningsmått kan användas för att prognostisera framtida räntabilitet på eget kapital och att placeringsstrategier baserade på offentligt tillgänglig årsredovisningsinformation har genererat en avkastning utöver vad som motiveras av olika mått på placeringsrisk. I studien observeras dock betydande tidsmässiga instabiliteter beträffande såväl möjligheterna att prognostisera framtida räntabilitet på eget kapital, som förekomsten av avkastning utöver vad som motiveras av placeringsrisk. / <p>Diss. Stockholm : Handelshögskolan, 2002</p>

財務報表記錄可說是企業經營績效良窳的反映指標，而其中所衍生出來的財務比率，向 來均是管理者、投資者進行企業診斷或未來經營績效預測的重要資訊來源。然而，相關 的研究發現，由於產業間經濟環境與市場結構特性的不同，所呈現出來的財務報表資訊 內涵亦將有所差別。因此，若進一步運用個別產業之報表資訊預測公司未來盈餘時，將 能夠提供產業間結果進行分析與比較的基礎。 如何自報表中獲取與公司經營績效相關之會計資訊，進而建構出優良的盈餘預測模式， 是近幾年來學者感興趣的研究課題之一。鑑於人工智慧之類神經網路系統擁有多項的特點，因此，對於盈餘預測會計資訊萃取的應用上，無非是提供了我們一個新的選擇途徑。 本研究即根據此項概念，以民國70年第一季至民國82年第三季為止共十五項大小產業之 股票上市公司財務報表以及股價報酬等資料作為研究樣本，進行盈餘預測模式的建構以 及投資超額報酬的計算。 進一步地說，本研究的內容可以分成三個部份，第一部份是以整體市場樣本為例，對類 神經網路主要參數如輸入變數組合、隱藏層節點數等進行調整及測試，以從中選取出盈 餘預測效果較佳之模式設定；在第二部份則是運用此一盈餘預測模式，分別對整體市場 以及紡織、電子電機 兩項產業樣本進行網路的訓練與測試，並根據模式所獲得之區別及 預測能力評估指標，探討不同產業特性樣本所建構的模式之間，其預測結果上的差異性 ；而第三部份則是利用各類產業模式預測結果的資訊，從利潤與風險兩種角度，定義"總 體"、"高利潤"、"低風險"、 "高利潤低風險"等四種不同類型投資策略，並以事件研究 法計算各項策略所能獲取之累積超額報酬，最後，則根據各策略之獲利績效，進行產業 間的分析比較，以找出本研究各類特定產業之最適投資策略。 本研究根據前述方式所進行的實驗研究中，獲得了以下三點結論： 一、類神經網路盈餘預測模式之建構 (一)以整體市場樣本為對象所進行之網路的測試中，發現模式整體區別能力大致介於五 到七成之間；而整體預測能力則介於四到六成之間。 (二)本研究所找出盈餘預測效果較佳之網路模式設定如下：1.輸入變數組合：單因子多變量變異數分析之22項顯著性財務比率 2.網路架構(輸入層-隱藏層-輸出層)：22-22-1 3.連結權數初始值設定範圍：-0.1~0.1 二、產業盈餘預測結果之分析 (一)整體而言，產業間模式測試結果的差異並不大，其中以紡織產業的模式區別及預測 能力最好(70%以上)，電子電機產業次之，而整體市場模式的結果均不及兩項單一性產業。 (二)模式預測能力穩定性方面，各產業於五個年度間預測率的波動大致還算穩定，其中 就紡織產業而言，其年度之間模式預測能力的差別不大，但電子電機產業年度間的變化 則要比前者來得明顯。 三、產業投資策略績效之分析 (一)各類型投資策略的整體結果中，紡織與電子電機兩項產業的獲利績效相當，且均要 比整體市場來得好，其中，紡織產業之"高利潤低風險"策略所獲得的累積超額報酬(43.28%) 更居全體之冠。 (二)本研究所找出之個別產業最適投資策略分別為： 1.整體市場：總體策略、低風險策略 2.紡織產業：高利潤低風險策略、高利潤策略 3.電子電機產業：高利潤低風險策略、低風險策略 / Financial Statements are very important information indicating performance of corporations. Managers and investors use financial ratios as vital indexes to evaluate and predict operating results of corporations, and make their decisions. ategy, and compute CAR for each investment strategies. At last, I analyze the investing results of the four strategies for individual industry. ANN ( Artificial Nerual Network) shoot a new direction on researching application of abstracting accounting information which can efficiently predict earnings. According to results of relative researches, financial statements from different industries present and implicate different accounting information. If we further apply ANN on financial statement information to predict earnings of corporations, we can use the results as bases of analyses and comparisons among industries. Because ANN model has many advantages, in this research, I use financial statements and return on stocks from corporations as researching samples to construct prediction models and compute CAR(Cumulative Abcdrmal Return) on investments. These samples are chosen from 15 different industries and covered from the first quarter of 1981 to the third quarter of 1993. This research consists of three parts: 22 financial ratios selected by MANOVA First, I use the general market samples to adjust and predict the vital parameters of ANN models, such as the selection of input variable, the number of hidden node, and finally pick better setups for the prediction model. Second, I use this model to train and test samples from the general market, the textile, and the electron industry, and research the variation of predicting results by different models made up different industries by means of evaluation indexes . Third, I use the results predicted by the three different industry models, inspect of risk and return, to define four types of investment strategies -- "the general", "the high return", "the low risk", and "the high return - low risk" strategy, and compute CAR for each investment strategies. At last, I analyze the investing results of the four strategies for individual industry. After researching, I find:s of the textile and electron industry are better than the general markets'. 1.The better setups of ANN predition models are :industries are: (1)the selection of input variable:the 22 financial ratios selected by MANOVA (2)the ANN model topology(input node - hidden node - output node):22-22-1 rategy (3)the range of initial connection weights:-0.1~0.1 return - low risk strategy 2.The analyses of results predicted by the three different industry models are: (1)the predicting abilities of the textile and electron industry are better than the general markets'. 3.The proper investment strategies of individual industries are: (1)the general market:the general and the low risk strategy (2)the textile industry:the high return and the high return - low risk strategy (3)the electron industry:the low risk and the high return - low risk strategy

文獻顯示存貨對於銷售和盈餘具有預測能力（Bernard and Noel 1991）。本文進一步探討比較後進先出法和國際會計準則允許之存貨計價方法所揭露之存貨，對於銷售和盈餘之預測能力。2003年發布之國際會計準則第二號公報「存貨」，禁止公司採用後進先出法衡量存貨，本研究擬觀察後進先出法和非後進先出法存貨對公司銷售與盈餘的預測能力是否有所差異。 本研究選取採用後進先出法並且揭露後進先出存貨準備之公司做為樣本，計算出樣本公司在國際會計準則規定下應有之存貨水準，測試與比較後進先出法之存貨與依國際會計準則揭露之存貨，孰者對銷售與盈餘之預測更具攸關性。實證結果顯示，後進先出存貨與國際會計準則存貨代理變數之實證結果並不顯著，顯示存貨在銷售與盈餘之預測迴歸模型中為一雜訊，存貨對銷售和盈餘之預測並不具有增額資訊，也說明存貨生產平穩理論與避免缺貨理論無法解釋存貨對銷售和盈餘預測之關聯性，因此無法判斷採用何種存貨計價方法所揭露之存貨，對銷售與盈餘較具預測能力。 / In economic literature, production smoothing model and stockout model address the predictability of inventory disclosure on sales and earnings. Based on these models, Bernard and Noel (1991) show that inventory disclosure predicts sales and earnings. This study further investigates and compares the predictability of the sales and earnings by inventory reported under last in, last out (LIFO) and that under International Accounting Standard 2 (IAS 2). Thus this study compares the predicting ability of inventory on sales and earnings under IFRS and non-IFRS. This study selects the companies adopting LIFO and disclosing LIFO reserve, calculates the inventory reported under IFRS, and determines the inventory’s ability to predict future sales and earnings under different inventory valuation methods. The empirical results show that the coefficients for the unexpected inventories under LIFO and IFRS are both statistically insignificant, suggesting that the unexpected inventories are merely noises in the models, and that the effects of production smoothing model and stockout model are not prevailed. Thus, it is difficult to determine which inventory valuation method can generate the inventory that leads to better sales and earnings prediction.