Index

• Absolute return attribution method
• Accounting data
• Accuracy
  • binary classification problems and
  • measurement of
• AdaBoost implementation
• Adaptable I/O System (ADIOS)
• Alternative data
• Amihud's lambda
• Analytics
• Annualized Sharpe ratio
• Annualized turnover, in backtesting
• Asset allocation
  • classical areas of mathematics used in
  • covariance matrix in
  • diversification in
  • Markowitz's approach to
  • Monte Carlo simulations for
  • numerical example of
  • practical problems in
  • quasi-diagonalization in
  • recursive bisection in
  • risk-based. See also Risk-based asset allocation approaches
  • tree clustering approaches to
• Attribution
• Augmented Dickey-Fuller (ADF) test. See also Supremum augmented Dickey-Fuller (SADF) test
• Average holding period, in backtesting
• Average slippage per turnover

• Backfilled data
• Backtesters
• Backtesting
  • bet sizing in
  • common errors in
  • combinatorial purged cross-validation (CPCV) method in
  • cross-validation (CV) for
  • customization of
  • definition of
  • “false discovery” probability and
  • flawless completion as daunting task in
  • general recommendations on
  • machine learning asset allocation and
  • purpose of
  • as research tool
  • strategy risk and
  • strategy selection in
  • synthetic data in
  • uses of results of
  • walk-forward (WF) method of
• Backtest overfitting
  • backtesters’ evaluation of probability of
  • bagging to reduce
  • combinatorial purged cross-validation (CPCV) method for
  • concerns about risk of
  • cross-validation (CV) method and
  • decision trees and proneness to
  • definition of
  • discretionary portfolio managers (PMs) and
  • estimating extent of
  • features stacking to reduce
  • general recommendations on
  • historical simulations in trading rules and
  • hyper-parameter tuning and
  • need for skepticism
  • optimal trading rule (OTR) framework for
  • probability of. See Probability of backtest overfitting (PBO)
  • random forest (RF) method to reduce
  • selection bias and
  • support vector machines (SVMs) and
  • trading rules and
  • walk-forward (WF) method and
• Backtest statistics
  • classification measurements in
  • drawdown (DD) and time under water (TuW) in
  • efficiency measurements in
  • general description of
  • holding period estimator in
  • implementation shortfall in
  • performance attribution and
  • performance measurements in
  • returns concentration in
  • runs in
  • run measurements in
  • time-weighted rate of returns (TWRR) in
  • timing of bets from series of target positions in
  • types of
• Bagging
  • accuracy improvement using
  • boosting compared with
  • leakage reduction using
  • observation redundancy and
  • overfitting reduction and
  • random forest (RF) method compared with
  • scalability using
  • variance reduction using
• Bars (table rows)
  • categories of
  • dollar bars
  • dollar imbalance bars
  • dollar runs bars
  • information-driven bars
  • standard bars
  • tick bars
  • tick imbalance bars
  • tick runs bars
  • time bars
  • volume bars
  • volume imbalance bars
  • volume runs bars
• Becker-Parkinson volatility algorithm
• Bet sizing
  • average active bets approach in
  • bet concurrency calculation in
  • budgeting approach to
  • dynamic bet sizes and limit prices in
  • holding periods and
  • investment strategies and
  • meta-labeling approach to
  • performance attribution and
  • predicted probabilities and
  • runs and increase in
  • size discretization in
  • strategy-independent approaches to
  • strategy's capacity and
• Bet timing, deriving
• Betting frequency
  • backtesting and
  • computing
  • implied precision computation and
  • investment strategy with trade-off between precision and
  • strategy risk and
  • targeting Sharpe ratio for
  • trade size and
• Bias
• Bid-ask spread estimator
• Bid wanted in competition (BWIC)
• big data analysis
• Bloomberg
• Boosting
  • AdaBoost implementation of
  • bagging compared with
  • implementation of
  • main advantage of
  • variance and bias reduction using
• Bootstrap aggregation. See Bagging
• Bootstraps, sequential
• Box-Jenkins analysis
• Broker fees per turnover
• Brown-Durbin-Evans CUSUM test

• Cancellation rates
• Capacity, in backtesting
• Chow-type Dickey-Fuller test
• Chu-Stinchcombe-White CUSUM test
• Classification models
• Classification problems
  • class weights for underrepresented labels in
  • generating synthetic dataset for
  • meta-labeling and
• Classification statistics
• Class weights
  • decision trees using
  • functionality for handling
  • underrepresented label correction using
• Cloud systems
• Combinatorially symmetric cross-validation (CSCV) method
• Combinatorial purged cross-validation (CPCV) method
  • algorithm steps in
  • backtest overfitting and
  • combinatorial splits in
  • definition of
  • examples of
• Compressed markets
• Computational Intelligence and Forecasting Technologies (CIFT) project
  • Adaptable I/O System (ADIOS) and
  • business applications developed by
  • Flash Crash of 2010 response and
  • mission of
• Conditional augmented Dickey-Fuller (CADF) test
• Correlation to underlying, in backtesting
• Corwin-Schultz algorithm
• Critical Line Algorithm (CLA)
  • description of
  • Markowitz's development of
  • Monte Carlo simulations using
  • numerical example with
  • open-source implementation of
  • practical problems with
• Cross-entropy loss (log loss) scoring
• Cross-validation (CV)
  • backtesting through
  • combinatorial purged cross-validation (CPCV) method in
  • embargo on training observations in
  • failures in finance using
  • goal of
  • hyper-parameter tuning with
  • k-fold
  • leakage in
  • model development and
  • overlapping training observations in
  • purpose of
  • purging process in training set for leakage reduction in
  • sklearn bugs in
• CUSUM filter
• CUSUM tests
• CV. See Cross-validation

• Data analysis
  • financial data structures and
  • fractionally differentiated features and
  • labeling and
  • sample weights and
• Data curators
• Data mining and data snooping
• Decision trees
• Decompressed markets
• Deflated Sharpe ratio (DSR)
• Deployment team
• Dickey-Fuller test
  • Chow type
  • supremum augmented (SADF)
• Discretionary portfolio managers (PMs)
• Discretization of bet size
• Diversification
• Dollar bars
• Dollar imbalance bars (DIBs)
• Dollar performance per turnover
• Dollar runs bars (DRBs)
• Downsampling
• Drawdown (DD)
  • definition of
  • deriving
  • example of
  • run measurements using
• Dynamic bet sizes

• Econometrics
  • financial Big Data analysis and
  • financial machine learning versus
  • HRP approach compared with
  • investment strategies based in
  • paradigms used in
  • substitution effects and
  • trading rules and
• Efficiency measurements
  • annualized Sharpe ratio and
  • deflated Sharpe ratio (DSR) and
  • information ratio and
  • probabilistic Sharpe ratio (PSR) and
  • Sharpe ratio (SR) definition in
• Efficient frontier
• Electricity consumption analysis
• Engle-Granger analysis
• Ensemble methods
  • boosting and
  • bootstrap aggregation (bagging) and
  • random forest (RF) method and
• Entropy features
  • encoding schemes in estimates of
  • financial applications of
  • generalized mean and
  • Lempel-Ziv (LZ) estimator in
  • maximum likelihood estimator in
  • Shannon's approach to
• ETF trick
• Event-based sampling
• Excess returns, in information ratio
• Execution costs
• Expanding window method
• Explosiveness tests
  • Chow-type Dickey-Fuller test
  • supremum augmented Dickey-Fuller (SADF) test

• Factory plan
• Feature analysts
• Feature importance
  • features stacking approach to
  • importance of
  • mean decrease accuracy (MDA) and
  • mean decrease impurity (MDI) and
  • orthogonal features and
  • parallelized approach to
  • plotting function for
  • random forest (RF) method and
  • as research tool
  • single feature importance (SFI) and
  • synthetic data experiments with
  • weighted Kendall's tau computation in
  • without substitution effects
  • with substitution effects
• Features stacking importance
• Filter trading strategy
• Finance
  • algorithmization of
  • human investors’ abilities in
  • purpose and role of
  • usefulness of ML algorithms in
• Financial data
  • alternative
  • analytics and
  • essential types of
  • fundamental
  • market
• Financial data structures
  • bars (table rows) in
  • multi-product series in
  • sampling features in
  • unstructured, raw data as starting point for
• Financial Information eXchange (FIX) messages
• Financial machine learning
  • econometrics versus
  • prejudices about use of
  • standard machine learning separate from
• Financial machine learning projects
  • reasons for failure of
  • structure by strategy component in
• Fixed-time horizon labeling method
• Fixed-width window fracdiff (FFD) method
• Flash crashes
  • class weights for predicting
  • “early warning” indicators in
  • high performance computing (HPC) tools and
  • signs of emerging illiquidity events and
• Flash Crash of 2010
• F1 scores
  • measurement of
  • meta-labeling and
• Fractional differentiation
  • data transformation method for stationarity in
  • earlier methods of
  • expanding window method for
  • fixed-width window fracdiff (FFD) method for
  • maximum memory preservation in
• Frequency. See Betting frequency
• Fundamental data
• Fusion collaboration project
• Futures
  • dollar bars and
  • ETF trick with
  • non-negative rolled price series and
  • single futures roll method with
  • volume bars and

• Gaps series, in single future roll method
• Global Investment Performance Standards (GIPS)
• Graph theory
• Grid search cross-validation

• Hasbrouck's lambda
• Hedging weights
• Herfindahl-Hirschman Index (HHI) concentration
• HHI indexes
  • on negative returns
  • on positive returns
  • on time between bets
• Hierarchical Data Format 5 (HDF5)
• Hierarchical Risk Parity (HRP) approach
  • econometric regression compared with
  • full implementation of
  • Monte Carlo simulations using
  • numerical example of
  • practical application of
  • quasi-diagonalization in
  • recursive bisection in
  • standard approaches compared with
  • traditional risk parity approach compared with
  • tree clustering approaches to
• High-frequency trading
• High-low volatility estimator
• High-performance computing (HPC)
  • ADIOS and
  • atoms and molecules in parallelization and
  • CIFT business applications and
  • cloud systems compared with
  • combinatorial optimization and
  • electricity consumption analysis using
  • Flash Crash of 2010 response and
  • fusion collaboration project using
  • global dynamic optimal trajectory and
  • hardware for
  • integer optimization approach and
  • multiprocessing and
  • objective function and
  • pattern-finding capability in
  • software for
  • streaming data analysis using
  • supernova hunting using
  • use cases for
  • vectorization and
• Holding periods
  • backtesting and
  • bet sizing and
  • estimating in strategy
  • optimal trading rule (OTR) algorithm with
  • triple-period labeling method and
• Hyper-parameter tuning
  • grid search cross-validation and
  • log loss scoring used with
  • randomized search cross-validation and

• Implementation shortfall statistics
• Implied betting frequency
• Implied precision computation
• Indicator matrix
• Information-driven bars (table rows)
  • dollar imbalance bars
  • dollar runs bars
  • purpose of
  • tick imbalance bars
  • tick runs bars
  • volume imbalance bars
  • volume runs bars
• Information ratio
• Inverse-Variance Portfolio (IVP)
  • asset allocation numerical example of
  • Monte Carlo simulations using
• Investment strategies
  • algorithmization of
  • bet sizing in
  • evolution of
  • exit conditions in
  • human investors’ abilities and
  • log loss scoring used with hyper-parameter tuning in
  • profit-taking and stop-loss limits in
  • risk in. See Strategy risk
  • structural breaks and
  • trade-off between precision and frequency in
  • trading rules and algorithms in
• Investment strategy failure probability
  • algorithm in
  • implementation of algorithm in
  • probabilistic Sharpe ratio (PSR) similarity to
  • strategy risk and

• K-fold cross-validation (CV)
  • description of
  • embargo on training observations in
  • leakage in
  • mean decrease accuracy (MDA) feature with
  • overlapping training observations in
  • purging process in training set for leakage reduction in
  • when used
• Kyle's lambda

• Labeling
  • daily volatility at intraday estimation for
  • dropping unnecessary or under-populated labels in
  • fixed-time horizon labeling method for
  • learning side and size in
  • meta-labeling and
  • quantamental approach using
  • triple-barrier labeling method for
• Labels
  • average uniqueness over lifespan of
  • class weights for underrepresented labels
  • estimating uniqueness of
• Lawrence Berkeley National Laboratory (LBNL, Berkeley Lab)
• Leakage, and cross-validation (CV)
• Leakage reduction
  • bagging for
  • purging process in training set for
  • sequential bootstraps for
  • walk-forward timefolds method for
• Lempel-Ziv (LZ) estimator
• Leverage, in backtesting
• Limit prices, in bet sizing
• Log loss scoring, in hyper-parameter tuning
• Log-uniform distribution
• Look-ahead bias

• Machine learning (ML)
  • finance and
  • financial machine learning separate from
  • HRP approach using
  • human investors and
  • prejudices about use of
• Machine learning asset allocation. See also Hierarchical Risk Parity (HRP) approach
  • Monte Carlo simulations for
  • numerical example of
  • quasi-diagonalization in
  • recursive bisection in
  • tree clustering approaches to
• Market data
• Markowitz, Harry
• Maximum dollar position size, in backtesting
• Maximum likelihood estimator, in entropy
• Mean decrease accuracy (MDA) feature importance
  • computed on synthetic dataset
  • considerations in working with
  • implementation of
  • single feature importance (SFI) and
• Mean decrease impurity (MDI) feature importance
  • computed on synthetic dataset
  • considerations in working with
  • implementation of
  • single feature importance (SFI) and
• Message Passing Interface (MPI)
• Meta-labeling
  • bet sizing using
  • code enhancements for
  • description of
  • dropping unnecessary or under-populated labels in
  • how to use
  • quantamental approach using
• Meta-strategy paradigm
• Microstructural features
  • alternative features of
  • Amihud's lambda and
  • bid-ask spread estimator (Corwin-Schultz algorithm) and
  • Hasbrouck's lambda and
  • high-low volatility estimator and
  • Kyle's lambda and
  • microstructural information definition and
  • probability of informed trading and
  • Roll model and
  • sequential trade models and
  • strategic trade models and
  • tick rule and
  • volume-synchronized probability of informed trading (VPIN) and
• Mixture of Gaussians (EF3M)
• Model development
  • cross-validation (CV) for
  • overfitting reduction and
  • single feature importance method and
• Modelling
  • applications of entropy to
  • backtesting in
  • cross-validation in
  • econometrics and
  • ensemble methods in
  • entropy features in
  • feature importance in
  • hyper-parameter tuning with cross-validation in
  • market microstructure theories and
  • three sources of errors in
• Monte Carlo simulations
  • machine learning asset allocation and
  • sequential bootstraps evaluation using
• Multi-product series
  • ETF trick for
  • PCA weights for
  • single future roll in

• National laboratories
• Negative (neg) log loss scores
  • hyper-parameter tuning using
  • measurement of
• Noise, causes of
• Non-negative rolled price series

• Optimal trading rule (OTR) framework
  • algorithm steps in
  • cases with negative long-run equilibrium in
  • cases with positive long-run equilibrium in
  • cases with zero long-run equilibrium in
  • experimental results using simulation in
  • implementation of
  • overfitting and
  • profit-taking and stop-loss limits in
  • synthetic data for determination of
• Options markets
• Ornstein-Uhlenbeck (O-U) process
• Orthogonal features
  • benefits of
  • computation of
  • implementation of
• Outliers, in quantitative investing
• Overfitting. See Backtest overfitting

• Parallelized feature importance
• PCA (see Principal components analysis)
• Performance attribution
• Plotting function for feature importance
• PnL (mark-to-market profits and losses)
  • ETF trick and
  • performance attribution using
  • as performance measurement
  • rolled prices for simulating
• PnL from long positions
• Portfolio construction. See also Hierarchical Risk Parity (HRP) approach
  • classical areas of mathematics used in
  • covariance matrix in
  • diversification in
  • entropy and concentration in
  • Markowitz's approach to
  • Monte Carlo simulations for
  • numerical example of
  • practical problems in
  • tree clustering approaches to
• Portfolio oversight
• Portfolio risk. See also Hierarchical Risk Parity (HRP) approach; Risk; Strategy risk
  • portfolio decisions based on
  • probability of strategy failure and
  • strategy risk differentiated from
• Portfolio turnover costs
• Precision
  • binary classification problems and
  • investment strategy with trade-off between frequency and
  • measurement of
• Predicted probabilities, in bet sizing
• Principal components analysis (PCA)
  • hedging weights using
  • linear substitution effects and
• Probabilistic Sharpe ratio (PSR)
  • calculation of
  • as efficiency statistic
  • probability of strategy failure, similarity to
• Probability of backtest overfitting (PBO)
  • backtest overfitting evaluation using
  • combinatorially symmetric cross-validation (CSCV) method for
  • strategy selection based on estimation of
• Probability of informed trading (PIN)
• Probability of strategy failure
  • algorithm in
  • implementation of algorithm in
  • probabilistic Sharpe ratio (PSR), similarity to
  • strategy risk and
• Profit-taking, and investment strategy exit
• Profit-taking limits
  • asymmetric payoff dilemma and
  • cases with negative long-run equilibrium and
  • cases with positive long-run equilibrium and
  • cases with zero long-run equilibrium and
  • daily volatility at intraday estimation points computation and
  • investment strategies using
  • learning side and size and
  • optimal trading rule (OTR) algorithm for
  • strategy risk and
  • triple-barrier labeling method for
• Purged K-fold cross-validation (CV)
  • grid search cross-validation and
  • hyper-parameter tuning with
  • implementation of
  • randomized search cross-validation and
• Python

• Quantamental approach
• Quantamental funds
• Quantitative investing
  • backtest overfitting in
  • failure rate in
  • meta-strategy paradigm in
  • quantamental approach in
  • seven sins of
• Quantum computing

• Random forest (RF) method
  • alternative ways of setting up
  • bagging compared with
  • bet sizing using
• Randomized search cross-validation
• Recall
  • binary classification problems and
  • measurement of
• Reinstated value
• Return attribution method
• Return on execution costs
• Returns concentration
• RF. See Random forest (RF) method
• Right-tail unit-root tests
• Risk. See also Hierarchical Risk Parity (HRP) approach; Strategy risk
  • backtest statistics for uncovering
  • entropy application to portfolio concentration and
  • liquidity and
  • ML algorithms for monitoring
  • PCA weights and
  • portfolio oversight and
  • profit taking and stop-loss limits and
  • structural breaks and
  • walk-forward (WF) approach and
• Risk-based asset allocation approaches
  • HRP approach comparisons in
  • structural breaks and
• Risk parity. See also Hierarchical Risk Parity (HRP) approach
  • HRP approach compared with traditional approach to
• Rolled prices
• Roll model

• Sample weights
  • average uniqueness of labels over lifespan and
  • bagging classifiers and uniqueness and
  • indicator matrix for
  • mean decrease accuracy (MDA) feature importance with
  • number of concurrent labels and
  • overlapping outcomes and
  • return attribution method and
  • sequential bootstrap and
  • time-decay factors and
• Sampling features
  • downsampling and
  • event-based sampling and
• Scalability
  • bagging for
  • sample size in ML algorithms and
• Scikit-learn (sklearn)
  • class weights in
  • cross-validation (CV) bugs in
  • grid search cross-validation in
  • labels and bug in
  • mean decrease impurity (MDI) and
  • neg log loss as scoring statistic and bug in
  • observation redundancy and bagging classifiers in
  • random forest (RF) overfitting and
  • support vector machine (SVM) implementation in
  • synthetic dataset generation in
  • walk-forward timefolds method in
• Selection bias
• Sequential bootstraps
  • description of
  • implementation of
  • leakage reduction using
  • Monte Carlo experiments evaluating
  • numerical example of
• Shannon, Claude
• Sharpe ratio (SR) in efficiency measurements
  • annualized
  • definition of
  • deflated (DSR)
  • information ratio and
  • probabilistic (PSR)
  • purpose of
  • targeting, for various betting frequencies
• Shorting, in quantitative investing
• Signal order flows
• Simulations, overfitting of
• Single feature importance (SFI)
• Single future roll
• Sklearn. See Scikit-learn
• Stacked feature importance
• Standard bars (table rows)
  • dollar bars
  • purpose of
  • tick bars
  • time bars
  • volume bars
• Stationarity
  • data transformation method to ensure
  • fractional differentiation applied to
  • fractional differentiation implementation methods for
  • integer transformation for
  • maximum memory preservation for
  • memory loss dilemma and
• Stop-loss, and investment strategy exit
• Stop-loss limits
  • asymmetric payoff dilemma and
  • cases with negative long-run equilibrium and
  • cases with positive long-run equilibrium and
  • cases with zero long-run equilibrium and
  • daily volatility computation and
  • fixed-time horizon labeling method and
  • investment strategies using
  • learning side and size and
  • optimal trading rule (OTR) algorithm for
  • strategy risk and
  • triple-barrier labeling method for
• Storytelling
• Strategists
• Strategy risk
  • asymmetric payouts and
  • calculating
  • implied betting frequency and
  • implied precision and
  • investment strategies and understanding of
  • portfolio risk differentiated from
  • probabilistic Sharpe ratio (PSR) similarity to
  • strategy failure probability and
  • symmetric payouts and
• Structural breaks
  • CUSUM tests in
  • explosiveness tests in
  • sub- and super-martingale tests in
  • types of tests in
• Sub- and super-martingale tests
• Supernova research
• Support vector machines (SVMs)
• Supremum augmented Dickey-Fuller (SADF) test
  • conditional ADF
  • implementation of
  • quantile ADF
• Survivorship bias
• SymPy Live
• Synthetic data
  • backtesting using
  • experimental results using simulation combinations with
  • optimal trading rule (OTR) framework using

• Tick bars
• Tick imbalance bars (TIBs)
• Tick rule
• Tick runs bars (TRBs)
• Time bars
  • description of
  • fixed-time horizon labeling method using
• Time-decay factors, and sample weights
• Time period, in backtesting
• Time series
  • fractional differentiation applied to
  • integer transformation for stationarity in
  • stationarity vs. memory loss dilemma in
• Time under water (TuW)
  • definition of
  • deriving
  • example of
  • run measurements using
• Time-weighted average price (TWAP)
• Time-weighted rate of returns (TWRR)
• Trading rules
  • investment strategies and algorithms in
  • optimal trading rule (OTR) framework for
  • overfitting in
• Transaction costs, in quantitative investing
• Tree clustering approaches, in asset allocation
• Triple-barrier labeling method
• Turnover costs

• Variance
  • boosting to reduce
  • causes of
  • ensemble methods to reduce
  • random forest (RF) method for
• Vectorization
• Volume bars
• Volume imbalance bars (VIBs)
• Volume runs bars (VRBs)
• Volume-synchronized probability of informed trading (VPIN)

• Walk-forward (WF) method
  • backtesting using
  • overfitting in
  • pitfalls of
  • Sharpe ratio estimation in
  • two key advantages of
• Walk-forward timefolds method
• Weighted Kendall's tau
• Weights. See Class weights; Sample weights