Trend-Following Backtester · Guide · backtest 한국어

equity · KR

HMM backtest

10 trend-following strategies were compared on the full daily history of HMM. 10 beat buy-and-hold CAGR; SMA 크로스오버 ranks first by Sharpe, matching the main backtester. The return/drawdown alternative is ADX / DI 방향성 (CAGR 8.5%, MDD -60.1%).

Data 2000-01-04 ~ 2026-07-31daily bars 6,644 (26.6 years)Costs 0.5%Venue XKRXMonthly snapshot 2026-07-31

Results by strategy

10 of 10 beat buy and hold (CAGR -2.5%, drawdown -99.3%).

Sharpe leader (same as main): SMA 크로스오버 — CAGR 11.6%, MDD -70.7%, Sharpe 0.49, 37% exposure.

Return/drawdown alternative: ADX / DI 방향성 — CAGR is 14.2%p higher than buy and hold, while drawdown improves by 28.6%p (CAGR 8.5%, MDD -60.1%, exposure 9%).

The alternative must retain 75% of buy-and-hold CAGR, improve drawdown by at least 20% and 5 percentage points, and stay below 95% exposure; the ranking weights return retention 60% and drawdown improvement 40%.

Strategy · parametersCAGRMDDSharpeCalmarTotalTradesExposure
SMA 크로스오버
fast=59, slow=277
11.6%-70.7%0.490.1618.6x1837%
ADX / DI 방향성
period=39, threshold=25
8.5%-60.1%0.480.148.7x189%
EMA 크로스오버
fast=55, slow=190
10.8%-74.3%0.470.1515.3x3235%
Donchian 채널 돌파
entryN=116, exitN=45
9.7%-71.9%0.450.1411.8x3024%
Keltner 채널 돌파
emaPeriod=59, atrPeriod=17, mult=4
8.3%-75.2%0.420.118.3x4217%
볼린저 밴드 돌파
n=53, k=2.5
6.7%-69.7%0.360.105.6x7423%
ROC 모멘텀
n=84, threshold=0.1
3.2%-91.6%0.260.032.3x25630%
MACD
fast=18, slow=34, signal=15
0.2%-96.0%0.220.001.0x32951%
Supertrend
period=25, mult=3.6
0.2%-93.8%0.200.001.1x14544%
Parabolic SAR
step=0.005, maxStep=0.16
-1.5%-93.1%0.17-0.020.7x19847%
Buy and hold-2.5%-99.3%0.26-0.030.5x1100%
0.1 1x 10x 2001 2005 2009 2013 2017 2021 2025 SMA 크로스오버Buy and hold
SMA 크로스오버: Solid strategy line versus dashed buy-and-hold line. Log scale.
0% -25% -50% -74% -99% 2001 2005 2009 2013 2017 2021 2025 Buy and holdSMA 크로스오버
SMA 크로스오버 maximum drawdown -70.7% (2010-10-06 → 2020-06-25), recovered after 3,743 days.

Rules of the top strategy

When the short average rises above the long average, it reads that as an uptrend starting.

Method and caveats

Past performance does not guarantee future returns. Parameters were selected on the same full-history sample using sharpe over 40 trials per strategy, so results contain in-sample bias. 0.5% trading costs are included; taxes, slippage and rejected orders are not.
Re-run the same setupAll asset reports

Read next