Equity · Global · ETFs, futures · Monthly Free

Empirical Asset Pricing via Machine Learning

Academic paper

to Open PDF]

AuthorsMeb Faber

Institute

Institut Mines-Télécom Business School
Cambria Investment Management

Paper linkhttps://papers.ssrn.com/sol3/papers.cfm?abstract_id=1585517

Strategy in a nutshell

The strategy uses machine learning to forecast one-month-ahead excess returns for 71 country stock markets from 1985–2021. Multiple models—including OLS, LASSO, RF, GBRT, and neural networks—predict returns using 88 market characteristics. Markets are ranked monthly, sorted into quintiles, and an equal-weighted long-short hedge portfolio is formed by going long on top-predicted markets and short on the lowest, with monthly rebalancing.

Economic rationale

The strategy’s success stems mainly from market mispricing rather than risk. Limits to arbitrage, illiquidity in small/emerging markets, and behavioral biases create exploitable anomalies. Combining multiple machine learning models enhances prediction stability and captures statistically significant abnormal returns..

Entry & exit

Entry

On each month, make_on_day sets target weights for EWZ, FXI, EWT, EWY, EIDO, THD, EPHE, ECH, EPOL, EZA (+2 more). Read engine.set_target_weights in the lab code for the exact signal.

Exit

A name is exited when its target weight is set to 0 (or the whole dict is empty = cash). Fills use 5 bps commission and 2 bps slippage.

Market data required

Needs at least 60 adjusted daily closes on EWZ, FXI, EWT, EWY, EIDO, THD (+6 more) before the first signal is ready. ETF columns print on exchange sessions only. The lab loads Yahoo Finance via backtest.data.load_daily_prices.

Symbols

EWZ, FXI, EWT, EWY, EIDO, THD, EPHE, ECH, EPOL, EZA, ARGT, TUR

Frequency

Monthly · Daily adjusted close; signals rebalance on the month change (last available session)

History needed

≥ 60 trading days per symbol (about 3 months), plus a few extra bars so the first signal can fire.

Fields

Adjusted close (Yahoo auto_adjust)
Session date index

Credentials

None for paper and the onsite lab
Broker or exchange API keys only for live orders

APIs & providers

Quant Buffet Data API

The lab calls load_daily_prices for the ASSETS panel of adjusted closes, then caches CSV so backtests replay the same series.

Provider: backtest.data.load_daily_prices
Auth: None for paper / sandbox backtests
Endpoints: load_daily_prices(ASSETS, start="2000-01-01")

API / docs → /en/docs/data-api

Yahoo Finance via yfinance

US equities and ETFs, adjusted OHLCV, free for research (rate limits apply).

Provider: yfinance → Yahoo Finance
Auth: None for public history
Endpoints: Ticker.history / yf.download(auto_adjust=True)

API / docs → github.com/ranaroussi/yfinance

Onsite backtest IDE

Quant Buffet native backtest IDE

Edit and run Quant Buffet Python for Empirical Asset Pricing via Machine Learning in the browser. Results update live with equity, drawdown, and metrics charts. Allowed: backtest.data, backtest.engine, backtest.metrics, numpy, pandas. Define ASSETS and make_on_day(prices). Shortcut: Ctrl+Enter. API docs →

IDE · 43 lines

"""
Quant Buffet interactive lab — empirical-asset-pricing-via-machine-learning
Allowed: backtest.data / backtest.engine / backtest.metrics, numpy, pandas.
Required: a module-level ASSETS list, and def make_on_day(prices) returning (on_day, ready)
"""
from __future__ import annotations

import numpy as np
import pandas as pd

from backtest.data import load_daily_prices
from backtest.engine import EngineConfig, PortfolioEngine
from backtest.metrics import compute_metrics

ASSETS = ['EWZ', 'FXI', 'EWT', 'EWY', 'EIDO', 'THD', 'EPHE', 'ECH', 'EPOL', 'EZA', 'ARGT', 'TUR']
REBALANCE_ONCE = False

def make_on_day(prices: pd.DataFrame):
    cols = [c for c in ASSETS if c in prices.columns]
    state = {"last": None, "done": False}

    def on_day(engine: PortfolioEngine, dt: pd.Timestamp) -> None:
        available = [s for s in cols if pd.notna(prices.at[dt, s])]
        if not available:
            return
        if REBALANCE_ONCE:
            if state["done"]:
                return
            state["done"] = True
            engine.set_target_weights(dt, {s: 1.0 / len(available) for s in available})
            return
        key = (dt.year, dt.month)
        if state["last"] == key:
            return
        state["last"] = key
        engine.set_target_weights(dt, {s: 1.0 / len(available) for s in available})

    counts = prices[cols].notna().sum(axis=1)
    need = max(1, len(cols) // 2)
    eligible = counts[counts >= need]
    ready = eligible.index.min() if not eligible.empty else None
    return on_day, ready

"""
Quant Buffet interactive lab — empirical-asset-pricing-via-machine-learning
Allowed: backtest.data / backtest.engine / backtest.metrics, numpy, pandas.
Required: a module-level ASSETS list, and def make_on_day(prices) returning (on_day, ready)
"""
from __future__ import annotations

import numpy as np
import pandas as pd

from backtest.data import load_daily_prices
from backtest.engine import EngineConfig, PortfolioEngine
from backtest.metrics import compute_metrics

ASSETS = ['EWZ', 'FXI', 'EWT', 'EWY', 'EIDO', 'THD', 'EPHE', 'ECH', 'EPOL', 'EZA', 'ARGT', 'TUR']
REBALANCE_ONCE = False

def make_on_day(prices: pd.DataFrame):
    cols = [c for c in ASSETS if c in prices.columns]
    state = {"last": None, "done": False}

def on_day(engine: PortfolioEngine, dt: pd.Timestamp) -> None:
        available = [s for s in cols if pd.notna(prices.at[dt, s])]
        if not available:
            return
        if REBALANCE_ONCE:
            if state["done"]:
                return
            state["done"] = True
            engine.set_target_weights(dt, {s: 1.0 / len(available) for s in available})
            return
        key = (dt.year, dt.month)
        if state["last"] == key:
            return
        state["last"] = key
        engine.set_target_weights(dt, {s: 1.0 / len(available) for s in available})

counts = prices[cols].notna().sum(axis=1)
    need = max(1, len(cols) // 2)
    eligible = counts[counts >= need]
    ready = eligible.index.min() if not eligible.empty else None
    return on_day, ready

Quant Buffet syntax cheat sheet (copy / insert)

Paste these fragments into the editor. The sandbox rejects QuantConnect, os, and network libraries.

Required imports

Only these libraries are allowed in the sandbox.

from __future__ import annotations

import numpy as np
import pandas as pd

from backtest.data import load_daily_prices
from backtest.engine import EngineConfig, PortfolioEngine
from backtest.metrics import compute_metrics

ASSETS list (whitelisted ETFs)

Module-level list. Tickers must be in the Quant Buffet whitelist.

ASSETS = ["SPY", "QQQ", "TLT", "GLD", "BIL"]

make_on_day contract

Must return (on_day, ready). on_day calls engine.set_target_weights.

def make_on_day(prices: pd.DataFrame):
    cols = [c for c in ASSETS if c in prices.columns]
    sma = prices[cols].rolling(200, min_periods=200).mean()
    state = {"last": None}

    def on_day(engine: PortfolioEngine, dt: pd.Timestamp) -> None:
        if sma.loc[dt].isna().all():
            return
        key = (dt.year, dt.month)
        if state["last"] == key:
            return
        state["last"] = key
        long = [
            s for s in cols
            if pd.notna(prices.at[dt, s]) and pd.notna(sma.at[dt, s])
            and prices.at[dt, s] > sma.at[dt, s]
        ]
        weights = {} if not long else {s: 1.0 / len(long) for s in long}
        engine.set_target_weights(dt, weights)

    ready = sma.dropna(how="all").index.min() if sma.notna().any().any() else None
    return on_day, ready

Set target weights

Weights should sum to about 1.0. Empty dict = 100% cash.

engine.set_target_weights(dt, {"SPY": 0.60, "BIL": 0.40})

Live backtest performance

CAGR

4.87%

Sharpe

0.32

Max DD

-63.05%

Vol

25.08%

Sortino

0.48

Beta

1.06

Run the backtest to populate charts.

Order records

Run the backtest to load every order fill alongside performance.

Export to your platform

Transform Quant Buffet lab code (ASSETS + make_on_day / PortfolioEngine) into native classes for a third-party IDE — then copy and paste.

Target platform

Run in: QuantConnect Cloud or LEAN CLI · QCAlgorithm with Equity securities and monthly rebalance.

Detected pattern: Custom / hybridAssets: EWZ, FXI, EWT, EWY, EIDO, THD, EPHE, ECH, EPOL, EZA, ARGT, TUR

# Generated from Quant Buffet → QuantConnect LEAN
# Strategy: Empirical Asset Pricing via Machine Learning
# Detected pattern: Custom / hybrid
# Source uses Quant Buffet lab APIs (ASSETS + make_on_day / PortfolioEngine).
# Review fees, data, and risk before live trading — educational export only.

from AlgorithmImports import *


class QuantBuffetExport(QCAlgorithm):
    def Initialize(self):
        self.SetStartDate(2010, 1, 1)
        self.SetCash(100000)
        tickers = ["EWZ", "FXI", "EWT", "EWY", "EIDO", "THD", "EPHE", "ECH", "EPOL", "EZA", "ARGT", "TUR"]
        self.symbols = []
        for t in tickers:
            if "-" in t:  # crypto proxy e.g. BTC-USD
                self.symbols.append(self.AddCrypto(t.replace("-USD", ""), Resolution.Daily).Symbol)
            else:
                self.symbols.append(self.AddEquity(t, Resolution.Daily).Symbol)
        self.Schedule.On(
            self.DateRules.MonthStart(self.symbols[0]),
            self.TimeRules.AfterMarketOpen(self.symbols[0], 30),
            self.Rebalance,
        )
        # Logic: Custom Quant Buffet logic — adapt the signal block to match your lab on_day().

    def Rebalance(self):
        # Pattern: custom — Custom Quant Buffet logic — adapt the signal block to match your lab on_day().
        # Default: equal-weight. Port your make_on_day weights here via SetHoldings.
        w = 1.0 / len(self.symbols) if self.symbols else 0.0
        for symbol in self.symbols:
            self.SetHoldings(symbol, w)

Exported code uses the platform’s native classes and libraries. Install dependencies in your third-party IDE, then run. Validate before live trading.