YATA

Yet Another Technical Analysis library

YaTa implements most common technical analysis methods and indicators.

It also provides you an interface to create your own indicators.

[dependencies]
yata = "0.2"

Some commonly used methods:

• Accumulation-distribution index;
• Cross / CrossAbove / CrossUnder;
• Derivative (differential)
• Highest / Lowest / Highest-Lowest Delta
• Hull moving average
• Integral (sum)
• Linear regression moving average
• Momentum
• Reverse points
• Simple moving average
• Weighted moving average
• Volume weighted moving average
• Exponential moving average family: EMA, DMA, TMA, DEMA, TEMA
• Symmetrically weighted moving average

And many others. See all

Some commonly used indicators:

• Average Directional Index;
• Awesome Oscillator;
• Bollinger Bands;
• Commodity Channel Index;
• Detrended Price Oscillator;
• Ease Of Movement;
• Elders Force Index;
• Envelopes;
• Fisher Transform;
• Ichimoku Cloud;
• Keltner Channels;
• Moving Average Convergence Divergence (MACD);
• Money Flow Index;
• Price Channel Strategy;
• Relative Strength Index (RSI);
• Stochastic Oscillator;
• Trix;
• Woodies CCI;

And many others. See all

Method usage example

use yata::prelude::*;
use yata::methods::EMA;

// EMA of length=3
let mut ema = EMA::new(3, 3.0).unwrap();

ema.next(3.0);
ema.next(6.0);

assert_eq!(ema.next(9.0), 6.75);
assert_eq!(ema.next(12.0), 9.375);

Indicator usage example

use yata::helpers::{RandomCandles, RegularMethods};
use yata::indicators::MACD;
use yata::prelude::*;
use std::convert::TryInto;

let mut candles = RandomCandles::new();
let mut macd = MACD::default();
macd.period3 = 4; // setting signal period MA to 4

macd.method1 = "sma".try_into().unwrap(); // one way of defining methods inside indicators

macd.method3 = RegularMethods::TEMA; // another way of defining methods inside indicators

let mut macd = macd.init(candles.first()).unwrap();

for candle in candles.take(10) {
	let result = macd.next(candle);

	println!("{:?}", result);
}

Benchmarks

Methods

• _w10 - method with window length=10
• _w100 - method with window length=100

test bench_adi_w10                   ... bench:          10 ns/iter (+/- 0)
test bench_adi_w100                  ... bench:          10 ns/iter (+/- 0)
test bench_cci_w10                   ... bench:          21 ns/iter (+/- 0)
test bench_cci_w100                  ... bench:          80 ns/iter (+/- 0)
test bench_conv_w10                  ... bench:          12 ns/iter (+/- 1)
test bench_conv_w100                 ... bench:         114 ns/iter (+/- 2)
test bench_cross                     ... bench:           6 ns/iter (+/- 3)
test bench_cross_above               ... bench:           4 ns/iter (+/- 0)
test bench_cross_under               ... bench:           4 ns/iter (+/- 0)
test bench_dema_w10                  ... bench:           9 ns/iter (+/- 0)
test bench_dema_w100                 ... bench:           9 ns/iter (+/- 0)
test bench_derivative_w10            ... bench:           2 ns/iter (+/- 0)
test bench_derivative_w100           ... bench:           2 ns/iter (+/- 0)
test bench_dma_w10                   ... bench:           3 ns/iter (+/- 0)
test bench_dma_w100                  ... bench:           3 ns/iter (+/- 0)
test bench_ema_w10                   ... bench:           7 ns/iter (+/- 1)
test bench_ema_w100                  ... bench:           7 ns/iter (+/- 0)
test bench_highest_index_w10         ... bench:           8 ns/iter (+/- 0)
test bench_highest_index_w100        ... bench:           9 ns/iter (+/- 0)
test bench_highest_lowest_delta_w10  ... bench:          12 ns/iter (+/- 0)
test bench_highest_lowest_delta_w100 ... bench:           5 ns/iter (+/- 0)
test bench_highest_w10               ... bench:           3 ns/iter (+/- 0)
test bench_highest_w100              ... bench:           4 ns/iter (+/- 0)
test bench_hma_w10                   ... bench:           8 ns/iter (+/- 0)
test bench_hma_w100                  ... bench:           8 ns/iter (+/- 0)
test bench_integral_w10              ... bench:           4 ns/iter (+/- 0)
test bench_integral_w100             ... bench:          10 ns/iter (+/- 0)
test bench_lin_reg_w10               ... bench:          11 ns/iter (+/- 0)
test bench_lin_reg_w100              ... bench:          11 ns/iter (+/- 0)
test bench_linear_volatility_w10     ... bench:           5 ns/iter (+/- 0)
test bench_linear_volatility_w100    ... bench:           5 ns/iter (+/- 0)
test bench_lowest_index_w10          ... bench:           7 ns/iter (+/- 4)
test bench_lowest_index_w100         ... bench:           4 ns/iter (+/- 0)
test bench_lowest_w10                ... bench:           3 ns/iter (+/- 0)
test bench_lowest_w100               ... bench:           4 ns/iter (+/- 0)
test bench_mean_abs_dev_w10          ... bench:           6 ns/iter (+/- 0)
test bench_mean_abs_dev_w100         ... bench:          77 ns/iter (+/- 0)
test bench_median_abs_dev_w10        ... bench:          43 ns/iter (+/- 1)
test bench_median_abs_dev_w100       ... bench:         288 ns/iter (+/- 1)
test bench_momentum_w10              ... bench:           4 ns/iter (+/- 0)
test bench_momentum_w100             ... bench:           4 ns/iter (+/- 0)
test bench_past_w10                  ... bench:           2 ns/iter (+/- 0)
test bench_past_w100                 ... bench:           2 ns/iter (+/- 0)
test bench_rate_of_change_w10        ... bench:           2 ns/iter (+/- 0)
test bench_rate_of_change_w100       ... bench:           2 ns/iter (+/- 0)
test bench_reverse_high_w10          ... bench:           5 ns/iter (+/- 0)
test bench_reverse_high_w100         ... bench:          10 ns/iter (+/- 0)
test bench_reverse_low_w10           ... bench:          12 ns/iter (+/- 0)
test bench_reverse_low_w100          ... bench:          10 ns/iter (+/- 0)
test bench_reverse_signal_w10        ... bench:          23 ns/iter (+/- 0)
test bench_reverse_signal_w100       ... bench:          20 ns/iter (+/- 0)
test bench_rma_w10                   ... bench:           7 ns/iter (+/- 0)
test bench_rma_w100                  ... bench:           3 ns/iter (+/- 0)
test bench_sma_w10                   ... bench:           2 ns/iter (+/- 0)
test bench_sma_w100                  ... bench:           2 ns/iter (+/- 0)
test bench_smm_w10                   ... bench:          14 ns/iter (+/- 0)
test bench_smm_w100                  ... bench:          44 ns/iter (+/- 4)
test bench_st_dev_w10                ... bench:          11 ns/iter (+/- 0)
test bench_st_dev_w100               ... bench:          10 ns/iter (+/- 0)
test bench_swma_w10                  ... bench:          12 ns/iter (+/- 1)
test bench_swma_w100                 ... bench:          12 ns/iter (+/- 0)
test bench_tema_w10                  ... bench:          11 ns/iter (+/- 0)
test bench_tema_w100                 ... bench:           5 ns/iter (+/- 0)
test bench_tma_w10                   ... bench:           4 ns/iter (+/- 0)
test bench_tma_w100                  ... bench:           4 ns/iter (+/- 0)
test bench_trima_w10                 ... bench:           3 ns/iter (+/- 0)
test bench_trima_w100                ... bench:           3 ns/iter (+/- 0)
test bench_vwma_w10                  ... bench:           3 ns/iter (+/- 0)
test bench_vwma_w100                 ... bench:           6 ns/iter (+/- 0)
test bench_wma_w10                   ... bench:           8 ns/iter (+/- 0)
test bench_wma_w100                  ... bench:           8 ns/iter (+/- 0)

Indicators

test bench_awsome_oscillator                   ... bench:          27 ns/iter (+/- 4)
test bench_detrended_price_oscillator          ... bench:           9 ns/iter (+/- 11)
test bench_indicator_aroon                     ... bench:          65 ns/iter (+/- 1)
test bench_indicator_average_directional_index ... bench:          69 ns/iter (+/- 0)

Current unsafe status

By default, there is no unsafe code in the crate. But you can optionally enable unsafe_performance feature throw you Cargo.toml or by --feature flag in your CLI.

usafe_perfomance enables some unsafe code blocks, most of them are unsafe access to a vector's elements. For some methods it may increase performance by ~5-10%.

Features

• serde - enables serde crate support;
• period_type_u16 - sets PeriodType to u16;
• period_type_u32 - sets PeriodType to u32;
• period_type_u64 - sets PeriodType to u64;
• value_type_f32 - sets ValueType to f32;
• unsafe_performance - enables optional unsafe code blocks, which may increase performance;

Rust version

YaTa library supports Rust stable except two things:

• You can't run benchmarks with it;
• cargo doc will generate docs without cross-links;

Suggestions

You are welcome to give any suggestions about implementing new indicators and methods.

Say thanks

If you like this library, and you want to say thanks, you can do it also by donating to bitcoin address 1P3gTnaTK9LKSYx2nETrKe2zjP4HMkdhvK