by Scott Juds 7 PM PST, 9-23-2019 SumGrowth Strategies

by Scott Juds – SumGrowth Strategies – Sept. 2019 1 Disclaimers • DO NOT BASE ANY INVESTMENT DECISION SOLELY UPON MATERIALS IN THIS PRESENTATION • Neither SumGrowth Strategies nor I are a registered investment advisor or broker-dealer. • This presentation is for educational purposes only and is not an offer to buy or sell securities. • This information is only educational in nature and should not be construed as investment advice as it is not provided in view of the individual circumstances of any particular individual. • Investing in securities is speculative. You may lose some or all of the money that is invested. • Past results of any particular trading system are not a guarantee of future performance. • Always consult with a registered investment advisor or licensed stock broker before investing.

2 Also Show Client Account with AD Risk as it should Be

Compare to S&P500

3 Also Show Client Account with AD Risk as it should Be

Compare to S&P500

4 5 Merlyn.AI Builds On (and does not have to re-discover) Other Existing Knowledge

It Doesn’t Have to Reinvent All of this From Scratch Formally Confirmed Momentum in Market Data

Merlyn.AI Accepts This Re-Discovery Not Necessary

Eugene Fama Kenneth French Nobel Prize, 2013 Dartmouth College “the premier market anomaly” that’s “above suspicion.”

Academic Paper - 2008: “Dissecting Anomalies” Proved Signal-to-Noise Ratio Controls the Probability of Making the Right Decision Claude Shannon National Medal of Science, 1966

Merlyn.AI Accepts This Re-Discovery Not Necessary Matched Filter Theory

Design for Optimum Signal-to-Noise Ratio J. H. Van Vleck Noble Prize, 1977 Think Outside of the Box

Merlyn.AI Accepts This Re-Discovery Not Necessary

Someplace to Start Designed for Performance Differential Signal Processing Removes Common Mode Noise (Relative Strength) Samuel H. Christie Royal Society 1836 Merlyn.AI Accepts This Re-Discovery Not Necessary 5 Years Full Span

Wheatstone Bridge Sectors Provide Power Strokes

Merlyn.AI Market Economic Cycle Cycle Accepts This Re-Discovery Not Necessary StormGuard-Armor Detect the Onset of Bad Markets

Know When the Market is Safe: Risk-On vs Risk-Off

Merlyn.AI Accepts This Re-Discovery Not Necessary StormGuard-Armor + Integrated Bear Market Strategies

Merlyn.AI Accepts This Re-Discovery Not Necessary These Guys Got us Here

Is There More? What About Selection Bias? Who Needs XLV-Healthcare and XLE-Energy?

.

. .

. Merlyn.AI Is a Genetic Algorithm Layered on Top of a Strategy

Genetic Algorithm on Top Why? To Evolve its Set of Funds Each Month

Why? To Remove Hindsight Selection Bias

Why? To Achieve Better Future Performance Artificial Intelligence Algorithms

Types of machine learning algorithms[edit] •Reinforcement learning Bayesian[edit] Semi-supervised learning[edit] •Almeida–Pineda recurrent backpropagation •Repeated incremental pruning to produce error reduction (RIPPER) Bayesian statistics Semi-supervised learning •ALOPEX •Rprop •Bayesian knowledge base •Active learning – special case of semi-supervised learning •Backpropagation •Rule-based machine learning •Naive Bayes Generative models •Bootstrap aggregating •Skill chaining •Gaussian Naive Bayes •Low-density separation •CN2 algorithm •Sparse PCA •Multinomial Naive Bayes •Graph-based methods •Constructing skill trees •State–action–reward–state–action •Averaged One-Dependence Estimators (AODE) •Co-training •Dehaene–Changeux model •Stochastic gradient descent •Bayesian Belief Network (BBN) •Transduction •Diffusion map •Structured kNN •Bayesian Network (BN) Deep learning[edit] •Dominance-based rough set approach •T-distributed stochastic neighbor embedding Decision tree algorithms[edit] Deep learning •Dynamic time warping •Temporal difference learning Decision tree algorithm •Deep belief networks •Error-driven learning •Wake-sleep algorithm •Decision tree •Deep Boltzmann machines •Evolutionary multimodal optimization •Weighted majority algorithm (machine l •Classification and regression tree (CART) •Deep Convolutional neural networks •Expectation–maximization algorithm •Iterative Dichotomiser 3 (ID3) •Deep Recurrent neural networks •FastICA •C4.5 algorithm •Hierarchical temporal memory •Forward–backward algorithm Supervised learning •C5.0 algorithm •Generative Adversarial Networks •GeneRec •AODE •Chi-squared Automatic Interaction Detection (CHAID) •Deep Boltzmann Machine (DBM) •Genetic Algorithm for Rule Set Production •Artificial neural network •Decision stump •Stacked Auto-Encoders •Growing self-organizing map •Association rule learning algorithms •Conditional decision tree Other machine learning methods and problems[edit] •HEXQ • Apriori algorithm •ID3 algorithm •Anomaly detection •Hyper basis function network • Eclat algorithm •Random forest •Association rules •IDistance •Case-based reasoning •SLIQ •Bias-variance dilemma •K-nearest neighbors algorithm •Gaussian process regression Linear classifier[edit] •Classification •Kernel methods for vector output •Gene expression programming Linear classifier • Multi-label classification •Kernel principal component analysis •Group method of data handling (GMDH) •Fisher's linear discriminant •Clustering •Leabra •Inductive logic programming •Linear regression •Data Pre-processing •Linde–Buzo–Gray algorithm •Instance-based learning •Logistic regression •Empirical risk minimization •Local outlier factor •Lazy learning •Multinomial logistic regression •Feature engineering •Logic learning machine •Learning Automata •Naive Bayes classifier •Feature learning •LogitBoost •Learning Vector Quantization •Perceptron •Learning to rank •Manifold alignment •Logistic Model Tree •Support vector machine •Occam learning •Minimum redundancy feature selection •Minimum message length (decision trees, decision graphs, etc.) Unsupervised learning[edit] •Online machine learning •Mixture of experts • Nearest Neighbor Algorithm Unsupervised learning •PAC learning •Multiple kernel learning • Analogical modeling •Expectation-maximization algorithm •Regression •Non-negative matrix factorization •Probably approximately correct learning (PAC) learning •Vector Quantization •Reinforcement Learning •Online machine learning •Ripple down rules, a knowledge acquisition methodology •Generative topographic map •Semi-supervised learning •Out-of-bag error •Symbolic machine learning algorithms •Information bottleneck method •Statistical learning •Prefrontal cortex basal ganglia working memory •Support vector machines Artificial neural networks[edit] •Structured prediction •PVLV •Random Forests Artificial neural network • Graphical models •Q-learning •Ensembles of classifiers •Feedforward neural network Logic learning machine • Bayesian network •Quadratic unconstrained binary optimization • Bootstrap aggregating (bagging) •Self-organizing map • Conditional random field (CRF) •Query-level feature • Boosting (meta-algorithm) Association rule learning[edit] • Hidden Markov model (HMM) •Quickprop •Ordinal classification Association rule learning •Unsupervised learning •Radial basis function network •Information fuzzy networks (IFN) •Apriori algorithm •VC theory •Randomized weighted majority algorithm •Eclat algorithm How SumGrowth Will Use AI To Perceive the environment and take action to maximize success.

Adaptively changing the algorithm FWPT: Forward Walk based on the past character of the Progressive Tuning Old data. Walks through out-of-sample data for its buy/sell decisions.

Employs Fuzzy Logic to evaluate a composite of 12 measures of the StormGuard - Armor Old market’s character to determine current investment safety.

Uses a Genetic Algorithm to evolve Merlyn.AI the candidate funds in a population FWPP: Forward Walk New Progressive Picking of momentum strategies to eradicate remnants of hindsight selection bias. (Note: Play Merlyn video from desktop now) Merlyn’s Magic Portfolio Merlyn’s Magic Portfolio ETF Tax Efficiency?

An Exchange In-Kind Is a Non-Taxable Event Originally Designed for Moving an Accounts to Another Brokerage

Market Makers Exchange a Basket of Stocks for ETF Shares Individuals Buy and Sell Only ETF Shares – It is Not Like Mutual Fund Ownership

Market Makers Return ETF Shares to the ETF Company and Receive an In-Kind Exchange for the Current Basket of Stocks. Thus Trades Occur Within the ETF Change the Basket, BUT the Investor Gets Long Term Tax Treatment. ETF Liquidity Myth Merlyn.AI Corp. Founded Jan 2019, Raised $2.5M, Exclusive License from SGS to Create & Market Merlyn ETFs

Solactive US Bank RBC Calculator Custodian Publisher Market Maker

Alpha Architect SGS Merlyn.AI ETF Advisor SectorSurfer SGS License Exemptive Relief NYSE AlphaDroid Investors Web Services MAI Indexes ETF Sponsor Compliance Quasar Marketing Distributor Cable CNBC Mktg. Approval Advisor Shares SEC FINRA G. AdWords Articles Bull-Rider Bear-Fighter Index MAI Bull-Rider Bear-Fighter Index MAI Indexes Solactive: Index Calculator & Publisher Bull-Rider Bear-Fighter Index Google Search – Stay Up To Date Bull-Rider Bear-Fighter Almost Found in the Partner Store

BRBF Portfolio Coming Soon

30 Bull-Rider Bear-Fighter www.AlphaDroid.com/InfoPages/Merlyn-BRBF.aspx Docs In Riskalyze Partner Store (2/8 pages)

31 AlphaDroid Portfolios Logistics in Riskalyze

32 Why Measure Risk If You’re not Going to Fix it?

by Scott Juds – SumGrowth Strategies – Sept. 2019 by Scott Juds SumGrowth Strategies 33