๐Ÿ”ฌ

4-Month Game Plan

Quantitative Research New Grad Prep

Structured month-by-month roadmap for landing a QR role at Two Sigma, D.E. Shaw, WorldQuant, Citadel, or similar. Each topic includes interview questions, key concepts, and hands-on projects.

Month 1 โ€” April 2026

Foundations: Statistics, Python & Time Series

๐Ÿ“

Weekly Game Plan

Week 1

Python / NumPy / Pandas โ€” no loops, pure vectors

  • MonRead McKinney Ch 4-5 (NumPy fundamentals, broadcasting). Do all exercises.
  • TueRead McKinney Ch 10-11 (Pandas time series, DatetimeIndex, resampling). Build a returns calculator from scratch.
  • WedImplement rolling Sharpe ratio, rolling beta, rolling volatility using only Pandas vectorized ops.
  • ThuSolve 5 Pandas problems on LeetCode/StrataScratch. Focus on groupby + window functions.
  • FriDownload SPY daily data via yfinance. Compute weekly/monthly/annual returns and plot them.
Week 2

Probability & Statistics โ€” the language of quant research

  • MonBlitzstein Stat 110 lectures 1-5 (counting, conditional probability, BLT). Do problem sets.
  • TueAll major distributions by hand โ€” derive E[X] and Var[X] for Normal, Poisson, Exponential, Binomial.
  • WedMLE derivation for Gaussian: maximize log-likelihood, show ฮผฬ‚=Xฬ„ and ฯƒฬ‚ยฒ=(1/n)ฮฃ(xแตข-Xฬ„)ยฒ.
  • ThuHypothesis testing: z-test, t-test, F-test, chi-squared. Know when to use each. Practice computing p-values.
  • FriMultiple testing: Bonferroni correction and Benjamini-Hochberg FDR. Simulate 1000 tests on noise, verify FDR at 5%.
Week 3

Linear Algebra & Regression โ€” the backbone of factor models

  • MonOLS derivation from scratch: ฮฒฬ‚ = (X'X)โปยนX'y. Implement in NumPy. Compare to sklearn.
  • TueGauss-Markov theorem: BLUE. Assumptions and what breaks when violated.
  • WedMulticollinearity (VIF), heteroskedasticity (White test), autocorrelation (Durbin-Watson). Fix each.
  • ThuPCA from scratch: center data, compute covariance matrix, eigendecomposition, project onto top k eigenvectors.
  • FriApply OLS to regress SPX sector returns on Fama-French 3 factors. Interpret ฮฑ, ฮฒ, Rยฒ, residuals.
Week 4

Time Series โ€” stationarity, autocorrelation, forecasting

  • MonStationarity: ADF test (null = unit root), KPSS test (null = stationary). Test VIX, SPY returns.
  • TueACF and PACF plots: interpret to identify AR(p) and MA(q) order. Practice on simulated data.
  • WedFit ARIMA(1,0,1) to SPY log-returns. Check residuals with Ljung-Box. Plot forecasts.
  • ThuCointegration: Engle-Granger test. Find two cointegrated ETFs (e.g. GLD/SLV). Build a pairs trade signal.
  • FriGARCH(1,1) on SPY returns โ€” volatility clustering. Forecast realized vol next week.

Topics & Interview Prep

Essential Library โ€” Buy These Now

Priority-ordered. "Must" = non-negotiable before any research interview.

๐Ÿ“–

Heard on the Street

Timothy Crack

Must

Classic quant interview prep. Solve every probability problem twice.

๐Ÿ“–

Expected Returns

Antti Ilmanen

Must

Factor investing bible. Required before any quant research interview.

๐Ÿ“–

Active Portfolio Management

Grinold & Kahn

Must

Defines IC, IR, breadth โ€” the language of every QR conversation.

๐Ÿ“–

Advances in Financial ML

Lopez de Prado

Must

Modern QR bible. Chapters 7, 8, 11 on CV, features, backtesting.

๐Ÿ“–

All of Statistics

Larry Wasserman

Must

Best statistics reference for research interviews. Dense but complete.

๐Ÿ“–

Practical Guide to Quant Interviews

Xinfeng Zhou

Must

Stochastic calc + probability + brain teasers. Buy and grind it.

๐Ÿ“–

Quantitative Equity Portfolio Mgmt

Chincarini & Kim

High

Barra-style factor models from theory to Python implementation.

๐Ÿ“–

Time Series Analysis

Hamilton

High

Graduate-level reference. Required for any time-series modeling discussion.

๐Ÿ“–

101 Formulaic Alphas

WorldQuant (free)

High

Free paper. 101 real alpha expressions. Study every single one.

Only works in the Electron app

<webview> is an Electron-only tag. Run npm run electron:dev to use this.

25:00Focus
0

25 min focus ยท 5 min break ยท long break every 4 sessions