Work with simple return, compound growth, inflation, fees, taxes, drawdowns, and volatility. You will calculate how small differences in cost and risk can change a portfolio over years.

Trace the shift from open-outcry markets and human analysts to index funds, electronic trading, robo-advisors, quantitative funds, and AI-assisted research. This history explains why today’s tools are powerful, fast, and easy to misuse.

Read income statements, balance sheets, cash flow statements, and key ratios as evidence about a business. You will pull facts from a company filing and separate business performance from market hype.

Value stocks using earnings, cash flow, dividends, comparable companies, and margin of safety. You will build a short valuation memo that an AI tool can help check but not decide for you.

Use bond prices, interest rates, credit risk, duration, and yield curves to judge fixed-income investments. You will compare a bond fund, a Treasury, and a high-yield bond using the risks that matter most.

Build portfolios with diversification, correlation, asset allocation, rebalancing, and risk tolerance. You will draft a simple investment policy statement that becomes the guardrail for later AI work.

Spot overconfidence, loss aversion, herd behavior, recency bias, and narrative traps. You will create a decision checklist that keeps both human emotion and AI output from steering the portfolio blindly.

Handle tables, tickers, dates, missing values, splits, dividends, total return, and benchmark data. You will clean a small price dataset and see why bad data creates false confidence.

Use averages, distributions, confidence intervals, regression, correlation, and base rates in plain investing situations. You will test whether a pattern is meaningful or just noise in a small sample.

Set up a practical toolchain with spreadsheets, Python notebooks, APIs, data files, charts, and versioned notes. You will build a small research notebook that records both the data and the reasoning behind an investment idea.

Train and test simple models such as linear regression, decision trees, random forests, and gradient boosting on financial examples. The focus is on what the model can and cannot claim, not on treating predictions as certainty.

Handle overfitting, leakage, survivorship bias, look-ahead bias, transaction costs, and benchmark choice. You will run a basic backtest and then break it on purpose to see how easy fake profits are to create.

Work with prices that arrive in order: trends, seasonality, volatility clustering, regime changes, and rolling forecasts. You will compare simple time-series methods with machine learning on the same market question.

Use mean-variance ideas, factor exposure, constraints, turnover limits, and scenario tests to improve a portfolio. You will compare an optimized portfolio with a simpler diversified portfolio and judge whether the added complexity is worth it.

Turn annual reports, earnings calls, analyst notes, news, and central bank statements into structured research inputs. You will use natural language tools to summarize documents while checking sources, context, and missing evidence.

Use retrieval-augmented generation to answer questions from filings, transcripts, policies, and your own notes with citations. You will build a small document library that reduces hallucinations and makes AI research auditable.

Use modern LLMs to draft research questions, compare companies, generate code, stress-test assumptions, and produce investment memos. You will practice prompt patterns that force the model to show sources, uncertainty, and alternative views.

Use satellite images, web traffic, job postings, app data, shipping data, social signals, and card-spending data without ignoring quality and legality. You will judge whether an alternative dataset truly adds information or only adds noise and cost.

Score companies and funds on value, quality, momentum, size, volatility, profitability, and other factors. You will build a screen, test its historical behavior, and check whether it fits a real portfolio goal.

Measure portfolio drawdowns, concentration, liquidity, leverage, currency exposure, inflation exposure, and stress scenarios. You will build a risk dashboard that warns you before a single story dominates the portfolio.

Connect models to trade sizing, order types, spreads, slippage, market impact, and execution timing. You will see why a strong signal can fail after costs and why most investors should avoid high-speed trading games.

Use reinforcement learning and simulation for allocation, execution, and trading research while respecting their limits. The chapter shows where these methods can help and why unstable markets make them dangerous without strong controls.

Compare robo-advisors, direct indexing, tax-loss harvesting, fractional shares, and automated rebalancing. You will evaluate when automation improves discipline and when it hides fees, assumptions, or conflicts.

Handle model drift, changing regimes, data outages, API changes, alerts, documentation, and scheduled reviews. You will create a maintenance plan so an AI-supported strategy does not quietly decay after launch.

Protect accounts, API keys, personal data, research files, and AI conversations from theft or misuse. You will also handle phishing, deepfakes, fake performance claims, and manipulated online investment content.

Apply fiduciary duty, suitability, Reg BI, SEC and FINRA rules, advertising limits, privacy duties, disclosures, and recordkeeping where they apply. Personal investors, creators, and professionals will see the line between research, education, and regulated advice.

Run a complete process from goal, policy, data collection, model choice, research memo, portfolio construction, trade plan, risk checks, documentation, and review schedule. This capstone produces a defensible AI-supported investment plan rather than a pile of disconnected predictions.

Map the roles around AI investing, including personal investor, analyst, quant researcher, portfolio manager, financial advisor, data engineer, risk manager, and compliance specialist. You will identify proof-of-skill projects, useful credentials, ethical contribution paths, and reliable ways to stay current.