Professor Khan clearly explains the financial theory required to understand the concept of a “nonzero risk-adjusted return".

Explaining the concept of accruals.

Performance of broker recommendations and reviews the research on the information anomalies related to estimate revisions and changes in analyst recommendations.

This chapter includes a good overview of the history of the anomaly, an update of performance results, and a clear discussion of the possible causes of the anomaly.

The chapter provides a concise overview of the anomalies related to financial ratios, investment growth, and distress risk, and discusses the Piotroski and Mohanram stock scoring systems that can be used by individual investors to outperform value or growth indexes.

Reviewing the research on stock buybacks, IPO, SEOs, and other stock purchase and stock issuance anomalies.

Providing an optimistic view from a professional of the possibility of generating excess returns from insider information.

Providing a good overview of how technical trading evolved from a backwater to become a major theme of respected academic anomaly research.

Providing an in-depth overview of the opportunities to earn returns using seasonal anomalies.

Concluding that value is a real anomaly but that size may not be an anomaly and discusses the basics of tactical asset allocation.

In order to systematically analyze earnings quality across a large set of firms, Sloan first had to standardize all the measuresto facilitate the comparison of firms with vastly different sizes. Sloan accomplished this by scaling earnings, accruals, and cash flows by total assets.

半强式有效市场假说认为价格已充分反应出所有已公开的有关公司营运前景的信息。这些信息有成交价、成交量、盈利资料、盈利预测值，公司管理状况及其它公开披露的财务信息等。假如投资者能迅速获得这些信息，股价应迅速作出反应。

信息在市场里面起着关键的作用，信息搜集可以对投资者对资产估值做出更好的决策作用，因此社会倾注了大量的资源（包括人力和物力）在证券分析上面。但是也许正是因为股票上有这么多的分析师分析研究，根据半强式有效市场假说，这些分析报告能否产生的价值是一个疑问。

已经有研究指出，分析师推荐调整已经反映了企业的各种行为了，所以股价已经priced in，对于大部分股票平均来说分析师预测已经不再推动股价波动，但是在一小部分股价没有反映出分析师预测的股票上，可以产生积极作用，因此我们可以根据这一现象去寻找那些未反映分析师预测的股票。

ratio of fundamental value is not positively associated with the optimism of analysyst recommendations. 有可能是因为分析师并只用了简单的估值模型，所以导致推荐的内容并不准确，因此可以通过residual-income model来做一个估值策略。

还有可能是因为分析师的观点并不中立，带有自己利益的偏见，那么有些推荐的股票并不一定能符合分析师预测。

因此，可以设置选股策略1：筛选没有立即反映分析师预测的股票；选股策略2：找出通过自己模型定价出与分析师识推荐的股票价格有差异的股票。

交易策略：每个季度初，把每只股票的分析师推荐平均以后排序，买入10%最乐观的股票，卖出10%最悲观的股票，每季节做一次rebalance.

Recommendation revisions are more valuable than recommendation levels.

Ball and Philip found that stocks with the largest variance in earnings also had the largest variance in stock price. Additionally, they found that stocks tended to move as long as three months in the direction of the earnings surprise.

分析师预测因子跟其它基本面因子结合，例如和高价格动量，高EP ratio，高BP ratio，低应收帐款，低成长，低资本支出几个因子结合，能提升策略表现效果。

跟踪能够持续预测的分析师的强烈买入信号的策略表现可能会比较好。

recommendations are more associated with less favorable value-to-price ratio companies.

IPO firms underperform the industry- and size-matched firms by 27.4%. Ritter (1991) concludes that the IPO issuing firms underperform their peer firms in both the buy-and-hold and the monthly rebalancing strategies. Ritter conjectures that this mispricing may be due to firms going public when investors are too optimistic about their prospects.

research has shown that firms that issued equity through an SEO, experience negative stock returns in the years following the equity issues.

market-timing: managers exploit their information advantage relative to outsiders to time their SEOs at opportune times when their stock is most likely to be overvalued.

straight debt issuances are associated with an average abnormal negative stock return of 14.3% for a period of 5 years, the evidence suggests that abnormal stock returns for the straight debt call sample is between 0.18% and 0.34% per month, depending on the benchmark used for a period of 5 years following the debt call event. As for convertible debt, the authors find mixed evidence following the calls of convertible debt.

Individual investors have been found to hold losers too long and sell winners too soon—called the disposition effect处置效应 (Shefrin and Statman 1985; Barber and Odean 2000; Grinblatt and Keloharju 2001), and only sell stocks that have experienced recent positive returns (Griffin, Harris, and Topaloglu 2003). Because the pain from the loss is more effective than the joy from the winning, it's hard for them to acknoledge the loss so that the loss trades are held longer than the winning trade.

The implication of such behavior is that individual investors are likely to be considered more as contrarian(逆势) investors and less as momentum-based investors.

These explanations typically fall into one of two categories: risk-based and investor behavior-based arguments.

The risk-based argument is based on the idea that high-risk investments should earn higher returns. Thus, it could be the case that momentum returns are merely compensation to investors for the inherent high risk of momentum portfolios or compensation for some unique risk associated with momentum investing.

However, as some of the aforementioned research has shown, it appears that momentum returns remain significant even after controlling for risk.

3-factor risk model fully explains the long-term reversals. after controlling for risk, returns from long-term reversal strategies become insignificant but momentum strategies remain profitable on a risk-adjusted basis.

so the reversal strategies can't fully explained by the risk theory.

• explanation of negative autocorrelation reversal and autocorrelation momentum.

Hong and Stein (1999) present a model of investor behavior based on the existence of two types of rational agents: “newswatchers” and “momentum traders.” In their model, newswatchers trade on fundamental information whereas momentum traders make trades based on past price movements. New, fundamental information diffuses gradually across the newswatchers and this causes prices to underreact and display positive autocorrelation (momentum). The momentum in prices incentivizes the momentum traders to enter the market; their simple trading strategies based on past prices eventually drive prices above fundamental value, leading prices to overreact and display negative autocorrelations over longer horizons (reversals) as prices move back to fundamental value.

Momentum returns can mostly be explained by changes in the business cycle.

初探A股季节性效应 民生证券首席策略分析师