Equity Management: The Art and Science of Modern Quantitative Investing (Bruce I. Jacobs; Kenneth N. Levy, 2018)

Review: Your complete guide for anything about equity - from history to current strategies to the right formulas to get the optimal portfolio. A long read but worth it. Rated: 9.5/10

Quantitative and measurement

• Alpha: measures the amount that the investment has returned in comparison to the market index or other broad benchmark that it is compared against
• Mean-variance-leverage optimisation (MVO): a quantitative tool that will allow you to make this allocation by considering the trade-off between risk and return
• Markowitz general mean-variance portfolio selection model (GPSM)
• Black-Scholes-Merton (BSM) model: a mathematical model for pricing an options contract
• Dividend discount model (DDM): valuing a company's stock price based on its future dividend 
• Information coefficient (IC): measure of the merit of a predicted value

Regularities measurement

• Low P/E: outperform high P/E stocks
• Small size: correlation with future performance
• Dividend yield: capital gain vs dividend on tax advantage
• Neglect: neglected stocks tends to outperform
• Low price: low-priced stocks to produce extra rewards
• Book/price: high book value to outperform
• Sales/price: superior to E/P
• Cashflow/price: superior to earnings in valuation
• Sigma: compensation for unsystematic risk
• Beta: measures the volatility of an investment. It is an indication of its relative risk
• Co-skewness: prefers positive skewness
• Earnings controversy: uncertain stocks produce superior returns
• Trends in analysts’ earnings estimates: upgrade stocks produce abnormal returns
• Earnings surprise: recent earnings surprises tend to produce abnormal returns
• The “earnings torpedo” effect: high expectations to have negative surprises
• Relative strength: market is not efficient
• Residual reversal: near-term relative price strength tends to reverse
• January: investors’ behaviour to be irrational

Methodology

• Seemingly unrelated regression (SUR): linear regression model that consists of several regression equations
• Generalized least squares (GLS): estimate the unknown parameters

Strategies

• Long/short equity: buying equities that are expected to increase in value and selling short equities that are expected to decrease in value
• Market-neutral: seeks to profit from both increasing and decreasing prices in one or more markets
• Client $10 initial funding → $9 to purchase stock long → $9 purchased long from prime broker (custodian) → $9 securities from stock lenders → $9 securities sold short → $9 proceeds from short sale → $9 collateral for borrowed stock → $1 liquidity buffer
• $10 account, $9 long, $1 cash / $9 short; proceeds posted as collateral → S&P 500 -15% → longs +33%, shorts +27% → spread = 33% - 27% → short rebate +5% → interest on cash +5% → returns +10.4%
• Equitized: adds a permanent stock index futures overlay, which makes profit or losses, depending on the movement of the market
• Hedge: offset potential losses or gains that may be incurred by a companion investment.
• 120-20 / 130-30: 130% of starting capital allocated to long positions and accomplishing this by taking in 30% of the starting capital from shorting stocks
• Momentum trading: buying and selling assets according to the recent strength of price trends

Statistic

• T-statistic: the ratio of the departure of the estimated value of a parameter from its hypothesized value to its standard error
• R-squared: a statistical measure of how close the data are to the fitted regression line
• Q-statistic: a test statistic output by either the Box-Pierce test or, in a modified version which provides better small sample properties
• Chi-squared test: statistical hypothesis tests that are distributed under the null hypothesis
• Mean error (ME): average of all the errors in a set
• Mean absolute error (MAE): difference between two continuous variables
• Root-mean-square error (RMSE): differences between values predicted by a model or an estimator and the values observed
• Theil U: Entropy coefficient - measure of nominal association
• Bayes's theorem: probability of an event, based on prior knowledge of conditions that might be related to the event
• Pearson correlation coefficient: measure of the linear correlation between two variables X and Y
• Monotonic regression: fitting a free-form line to a sequence of observations under the following constraints

Other Mathematics

• Karush–Kuhn–Tucker (KKT): first derivative tests for a solution in nonlinear programming to be optimal, provided that some regularity conditions are satisfied

Risk measurement

• BARRA model: measure the overall risk associated with a security relative to the market
• Sharpe: performance of an investment compared to a risk-free asset (beta, yield, size, bond beta, alpha)
• Information ratio (IR): performance of an investment compared to a benchmark index, after adjusting for its additional risk

Size effect: returns are inversely related a firm’s size

• Size and transaction cost: overstatement because of the bid/ask effect
• Size and risk management: bias in beta leads to overstatement
• Size and risk premiums: more risk to estimate small firm as there are less information
• Size and other cross-sectional effect: such as low P/E
• Size and calendar effects: more prone to calendar anomalies

Notable firms

• Long-term capital management (LTCM): hedge fund in Greenwich
• MSCI Barra
• Jacobs levy markowitz market simulator (JLMSim)

Terms

• Efficient market hypothesis (EMH): asset prices reflect all available information
• Capital asset pricing model (CAPM): required rate of return of an asset
• Arbitrage pricing theory (APT): asset's returns can be predicted using the linear relationship between the asset’s expected return and a number of macroeconomic variables that capture systematic risk
• Multi-factor model: a financial model that employs multiple factors in its calculations
• Disentangling and purify
• Leverage aversion: highest risk-adjusted return is achieved not by the market but, rather, by a portfolio that overweights safer assets
• Risk parity: portfolio management which focuses on allocation of risk, usually defined as volatility, rather than allocation of capital
• Efficient frontier: optimal portfolios that offer the highest expected return for a defined level of risk given level of expected return
• Residential mortgage-backed securities (RMBS): debt-based security (similar to a bond), backed by the interest paid on loans for residences
• Asset-backed commercial paper (ABCP): commercial paper that is collateralized by other financial assets
• Structured investment vehicle (SIV): a pool of investment assets that attempts to profit from credit spreads between short-term debt and long-term structured finance products
• Collateralized debt obligations (CDO): repackage individual loans into a product sold to investors on the secondary market
• Credit default swap (CDS): a financial swap agreement that the seller of the CDS will compensate the buyer in the event of a debt default
• Loan-to-value (LTV) ratio: the ratio of a loan to the value of an asset purchased

Interrelated return effects

• Earnings surprise effect
• Neglected-firm effect
• Return-reversal effect
• Small-size effect
• January effect
• Low-P/E effect
• Earnings-revision effect
• Book/price effect
• Low-price effect
• Yield effect 

Human tend to delay bad news

• Time-of-the-day effect: anomaly before close-of-day
• Day-of-the-week effect: weekend reporting causes bad Friday-to-Monday return
• Week-of-the-month effect: reporting later in the month
• Turn-of-the-month effect: stock prices to increase during the last two days and the first three days of each month
• Holiday Effect: stock market to gain on the final trading day before an exchange-mandated long weekend
• Late reporter anomaly: late announcements are often negative and cause a price decline

Technical indicators

• Vector autoregression (VAR): stochastic process model used to capture the linear interdependencies among multiple time series
• Vector autoregression moving-average (VARMA): overcome the overfitting problem inherent in VAR
• Bayesian vector autoregression (BVAR): uses Bayesian methods to estimate a VAR
• Autoregressive integrated moving average (ARIMA): better understand the data set or to predict future trends

Calculations

• Consensus trend = (current consensus mean - 1 month ago consensus mean) / price
• Flash trend = (current 6-week flash mean - 1 month ago 6-week flash mean) / price
• Stock return = consensus predictor + (flash - consensus predictor) + controversy + neglect
• Neglect = -log (1 + number of fiscal year 1 analysts)
• Portfolios
• Various utility functions
• Neutral portfolios
• Dollar-neutral portfolios
• Beta-neutral portfolios
• Optimal equitisation
• Weighted average market capitalization (WACW): stock market index whose components are weighted according to the total market value of their outstanding shares
• Critical line algorithm (CLA) is a portfolio optimization routine uses the weights that maximize the sharpe ratio, but the algorithm produces several sets of weights along the efficient frontier curve