A lightweight, end-to-end quant research project that builds and backtests multiple systematic strategies from scratch using daily market data.
Focus: cross-sectional signals, realistic backtesting, transaction costs, and portfolio construction.
Ranks assets by volatility-normalized rolling returns and forms a market-neutral portfolio:
- Long top quantile of momentum scores
- Short bottom quantile
- Monthly rebalancing
- Includes transaction costs (commission + slippage)
Runs a daily cross-sectional regression (market factor model) to estimate “expected” returns and trades residual mean reversion:
- Compute residuals = actual − predicted
- Long most negative residuals (undershoot)
- Short most positive residuals (overshoot)
- Monthly rebalancing + transaction costs
Builds a breakout-style signal using a rolling price channel and converts it into a cross-sectional score:
- Score is derived from position inside the rolling high/low channel
- Long top quantile, short bottom quantile (market-neutral)
- Monthly rebalancing + transaction costs
Combines strategy return streams using dynamic inverse-volatility weights:
- Compute rolling vol per strategy
- Allocate weights ∝ 1/vol
- Portfolio rebalanced daily (weights applied with a 1-day lag)
- Optional strategy-level turnover costs
A lightweight, end-to-end quant research project that builds and backtests multiple systematic strategies from scratch using daily market data.
Focus: cross-sectional signals, realistic backtesting, transaction costs, and portfolio construction.
Ranks assets by volatility-normalized rolling returns and forms a market-neutral portfolio:
- Long top quantile of momentum scores
- Short bottom quantile
- Monthly rebalancing
- Includes transaction costs (commission + slippage)
Runs a daily cross-sectional regression (market factor model) to estimate “expected” returns and trades residual mean reversion:
- Compute residuals = actual − predicted
- Long most negative residuals (undershoot)
- Short most positive residuals (overshoot)
- Monthly rebalancing + transaction costs
Builds a breakout-style signal using a rolling price channel and converts it into a cross-sectional score:
- Score is derived from position inside the rolling high/low channel
- Long top quantile, short bottom quantile (market-neutral)
- Monthly rebalancing + transaction costs
Combines strategy return streams using dynamic inverse-volatility weights:
- Compute rolling vol per strategy
- Allocate weights ∝ 1/vol
- Portfolio rebalanced daily (weights applied with a 1-day lag)
- Optional strategy-level turnover costs
A lightweight, end-to-end quant research project that builds and backtests multiple systematic strategies from scratch using daily market data.
Focus: cross-sectional signals, realistic backtesting, transaction costs, and portfolio construction.
Ranks assets by volatility-normalized rolling returns and forms a market-neutral portfolio:
- Long top quantile of momentum scores
- Short bottom quantile
- Monthly rebalancing
- Includes transaction costs (commission + slippage)
Runs a daily cross-sectional regression (market factor model) to estimate “expected” returns and trades residual mean reversion:
- Compute residuals = actual − predicted
- Long most negative residuals (undershoot)
- Short most positive residuals (overshoot)
- Monthly rebalancing + transaction costs
Builds a breakout-style signal using a rolling price channel and converts it into a cross-sectional score:
- Score is derived from position inside the rolling high/low channel
- Long top quantile, short bottom quantile (market-neutral)
- Monthly rebalancing + transaction costs
Combines strategy return streams using dynamic inverse-volatility weights:
- Compute rolling vol per strategy
- Allocate weights ∝ 1/vol
- Portfolio rebalanced daily (weights applied with a 1-day lag)
- Optional strategy-level turnover costs
A lightweight, end-to-end quant research project that builds and backtests multiple systematic strategies from scratch using daily market data.
Focus: cross-sectional signals, realistic backtesting, transaction costs, and portfolio construction.
Ranks assets by volatility-normalized rolling returns and forms a market-neutral portfolio:
- Long top quantile of momentum scores
- Short bottom quantile
- Monthly rebalancing
- Includes transaction costs (commission + slippage)
Runs a daily cross-sectional regression (market factor model) to estimate “expected” returns and trades residual mean reversion:
- Compute residuals = actual − predicted
- Long most negative residuals (undershoot)
- Short most positive residuals (overshoot)
- Monthly rebalancing + transaction costs
Builds a breakout-style signal using a rolling price channel and converts it into a cross-sectional score:
- Score is derived from position inside the rolling high/low channel
- Long top quantile, short bottom quantile (market-neutral)
- Monthly rebalancing + transaction costs
Combines strategy return streams using dynamic inverse-volatility weights:
- Compute rolling vol per strategy
- Allocate weights ∝ 1/vol
- Portfolio rebalanced daily (weights applied with a 1-day lag)
- Optional strategy-level turnover costs