kingchenc
75eefbbd08
The strategy_* examples were only syntax-smoked in CI, never run, which hid two classes of problem: 1. Python strategy_macd_adx / strategy_bollinger_squeeze passed three separate arguments to the candle indicators ADX/ATR, whose .update() takes a single candle — a TypeError at runtime — and read the ADX tuple at index 0 (plus_di) instead of 2 (adx). Both fixed. 2. The Go / C# / R / Java strategies defaulted to synthetic data and used a different (annualised) one-line summary, so they printed wildly different numbers from the Rust/Python/Node/C/WASM suite. Rewrite them to the shared per-trade backtest (load the bundled BTCUSDT CSV by default, same entry/exit logic, same print_summary output). All nine runnable bindings now print byte-identical backtest summaries on the same data (MACD+ADX 246 trades / -47.19%, RSI 37 / -17.84%, Bollinger 1 / -7.82%), verified by diffing each language's output against the Python reference. WASM shares the same logic and bundled dataset (browser-rendered).
70 lines
2.5 KiB
R
70 lines
2.5 KiB
R
# Strategy example: Bollinger-squeeze breakout with an ATR(14) trailing stop.
|
|
#
|
|
# Enters long when Bollinger bandwidth makes a new SQUEEZE_LOOKBACK low (a
|
|
# volatility squeeze) and price closes above the upper band; exits on an ATR(14)
|
|
# trailing stop or when the upper band falls back below the entry. 0.1% fees per
|
|
# trade. The R counterpart of examples/python/strategy_bollinger_squeeze.py,
|
|
# printing the same summary. Uses the checked-in examples/data/btcusdt-1d.csv
|
|
# dataset (pass a CSV path to override).
|
|
suppressPackageStartupMessages(library(wickra))
|
|
source("_common.R")
|
|
|
|
FEE <- 0.001
|
|
ATR_STOP_MULT <- 2.0
|
|
SQUEEZE_LOOKBACK <- 180
|
|
|
|
args <- commandArgs(trailingOnly = TRUE)
|
|
bars <- if (length(args) >= 1) load_ohlcv_csv(args[1]) else bundled_candles("btcusdt-1d.csv")
|
|
|
|
opens <- bars$open; highs <- bars$high; lows <- bars$low
|
|
closes <- bars$close; vols <- bars$volume; ts <- bars$timestamp
|
|
n_bars <- length(closes)
|
|
|
|
bb <- BollingerBands(20, 2.0); atr <- Atr(14)
|
|
in_pos <- FALSE; entry_price <- 0; stop_level <- 0
|
|
closed <- numeric(0); equity <- 1; equity_curve <- numeric(n_bars)
|
|
bw_window <- numeric(0)
|
|
|
|
for (i in seq_len(n_bars)) {
|
|
band <- update(bb, closes[i])
|
|
atr_value <- update(atr, opens[i], highs[i], lows[i], closes[i], vols[i], ts[i])
|
|
price <- closes[i]
|
|
equity_curve[i] <- if (in_pos) equity * (price / entry_price) else equity
|
|
if (is.na(band[["middle"]]) || !is.finite(atr_value)) next
|
|
|
|
middle <- band[["middle"]]
|
|
if (abs(middle) <= 1e-12) next
|
|
upper <- band[["upper"]]; lower <- band[["lower"]]
|
|
bandwidth <- (upper - lower) / middle
|
|
bw_window <- c(bw_window, bandwidth)
|
|
if (length(bw_window) > SQUEEZE_LOOKBACK) {
|
|
bw_window <- bw_window[(length(bw_window) - SQUEEZE_LOOKBACK + 1):length(bw_window)]
|
|
}
|
|
if (length(bw_window) < SQUEEZE_LOOKBACK) next
|
|
min_bw <- min(bw_window)
|
|
|
|
if (in_pos) {
|
|
if (price < stop_level || upper < entry_price) {
|
|
trade_ret <- price / entry_price - 1
|
|
closed <- c(closed, trade_ret)
|
|
equity <- equity * (1 + trade_ret) * (1 - FEE)
|
|
in_pos <- FALSE
|
|
}
|
|
} else {
|
|
is_new_low <- abs(bandwidth - min_bw) < 1e-12
|
|
if (is_new_low && price > upper) {
|
|
entry_price <- price; stop_level <- price - ATR_STOP_MULT * atr_value
|
|
equity <- equity * (1 - FEE); in_pos <- TRUE
|
|
}
|
|
}
|
|
}
|
|
|
|
if (in_pos) {
|
|
trade_ret <- closes[n_bars] / entry_price - 1
|
|
closed <- c(closed, trade_ret)
|
|
equity <- equity * (1 + trade_ret) * (1 - FEE)
|
|
}
|
|
|
|
print_summary("Bollinger Squeeze Breakout (1d, BTCUSDT)",
|
|
closes[1], closes[n_bars], n_bars, closed, equity, equity_curve)
|