NVI — Negative Volume Index¶
Tracks price changes on days when volume decreases, based on the theory that smart money trades on quiet days.
Inputs: [real, volume] | Options: [] | Outputs: [nvi]
Basic¶
use tulip_rs::indicators::nvi::{Nvi, Indicator, TIndicatorState};
let close = vec![81.59, 81.06, 82.87, 83.00, 83.61,
83.15, 82.84, 83.99, 84.55, 84.36_f64];
let volume = vec![1200.0, 1400.0, 1100.0, 1600.0, 1300.0,
900.0, 1500.0, 1800.0, 1000.0, 1700.0_f64];
let inputs = [close.as_slice(), volume.as_slice()];
let (outputs, mut state) = Nvi::indicator(&inputs, &[], None).unwrap();
println!("{:?}", outputs[0]); // NVI values
// State continuation — feed new bars without reprocessing history
let new_close = vec![85.00_f64];
let new_volume = vec![850.0_f64];
let continued = state.batch_indicator(
&[new_close.as_slice(), new_volume.as_slice()],
None,
).unwrap();
println!("{:?}", continued[0]);
#include "tulip_rs_ffi.h"
const double close[] = {81.59, 81.06, 82.87, 83.00, 83.61, 83.15, 82.84, 83.99, 84.55, 84.36,
85.53, 86.54, 86.89, 87.77, 87.29};
const double volume[] = {5653100.0, 6447400.0, 7690900.0, 3831400.0, 4455100.0, 3798000.0,
3936200.0, 4732000.0, 4841300.0, 3915300.0, 6830800.0, 6694100.0,
5293600.0, 7985800.0, 4807900.0};
const double *inputs[NVI_INPUTS] = {close, volume};
const double options[NVI_OPTIONS] = {}; // NVI has no options (OPTIONS=0)
/* Full computation */
CIndicatorResult r = nvi_indicator(inputs, 15, options, NULL, 0);
/* r.outputs[0] -> the NVI series, length r.output_lens[0] */
tulip_ffi_result_free(r);
nvi_state_free(r.state);
/* Partial computation + batch continuation */
CIndicatorResult pr = nvi_indicator(inputs, 10, options, NULL, 0);
const double *rest_inputs[NVI_INPUTS] = {close + 10, volume + 10};
CBatchResult br = nvi_batch(pr.state, rest_inputs, 5, NULL, 0);
/* br.outputs[0] -> NVI values for the last 5 bars */
tulip_ffi_batch_result_free(br);
tulip_ffi_result_free(pr);
nvi_state_free(pr.state);
import "github.com/me60732/tulip_rs_go/indicators"
close := []float64{81.59, 81.06, 82.87, 83.00, 83.61, 83.15, 82.84, 83.99, 84.55, 84.36,
85.53, 86.54, 86.89, 87.77, 87.29}
volume := []float64{5653100.0, 6447400.0, 7690900.0, 3831400.0, 4455100.0, 3798000.0,
3936200.0, 4732000.0, 4841300.0, 3915300.0, 6830800.0, 6694100.0,
5293600.0, 7985800.0, 4807900.0}
options := []float64{} // no options (NVI has zero options)
// Full computation — Rows are zero-copy views, valid until Close.
res, st, _ := indicators.Nvi.Indicator(close, volume, options, nil)
fmt.Println(res.Rows[0]) // NVI values
res.Close()
st.Close()
// Partial computation + state continuation.
res2, st2, _ := indicators.Nvi.Indicator(close[:10], volume[:10], options, nil)
res2.Close() // outputs consumed or closed; state stays live
batch, _ := st2.Batch(close[10:], volume[10:], nil)
fmt.Println(batch.Rows[0]) // continued NVI values
batch.Close()
st2.Close()
import org.tuliprs.*;
import org.tuliprs.indicators.Nvi;
double[] close = {81.59, 81.06, 82.87, 83.00, 83.61, 83.15, 82.84, 83.99, 84.55, 84.36,
85.53, 86.54, 86.89, 87.77, 87.29};
double[] volume = {5653100.0, 6447400.0, 7690900.0, 3831400.0, 4455100.0, 3798000.0,
3936200.0, 4732000.0, 4841300.0, 3915300.0, 6830800.0, 6694100.0,
5293600.0, 7985800.0, 4807900.0};
double[] options = {}; // no options
// Full computation — output rows are zero-copy views, valid until close().
Outcome oc = Nvi.indicator(new double[][] {close, volume}, options);
try (Result res = oc.result(); State st = oc.state()) {
System.out.println(java.util.Arrays.toString(res.toDoubleArray(0))); // NVI values
}
// Partial computation + state continuation.
int n = 10;
Outcome p = Nvi.indicator(new double[][] {
java.util.Arrays.copyOfRange(close, 0, n),
java.util.Arrays.copyOfRange(volume, 0, n)}, options);
try (Result pr = p.result(); State st = p.state()) {
Result br = st.batch(new double[][] {
java.util.Arrays.copyOfRange(close, n, 15),
java.util.Arrays.copyOfRange(volume, n, 15)});
try (br) {
System.out.println(java.util.Arrays.toString(br.toDoubleArray(0))); // continued NVI
}
}
import numpy as np
import tulip_rs
close = np.array([81.59, 81.06, 82.87, 83.00, 83.61,
83.15, 82.84, 83.99, 84.55, 84.36], dtype=np.float64)
volume = np.array([1200.0, 1400.0, 1100.0, 1600.0, 1300.0,
900.0, 1500.0, 1800.0, 1000.0, 1700.0], dtype=np.float64)
outputs, state = tulip_rs.indicators.nvi.indicator([close, volume], [])
print(outputs[0]) # NVI values
# State continuation
new_close = np.array([85.00], dtype=np.float64)
new_volume = np.array([850.0], dtype=np.float64)
continued = state.batch_indicator([new_close, new_volume])
print(continued[0])
import * as ti from 'tulip-rs-node';
const close = Float64Array.from([81.59, 81.06, 82.87, 83.00, 83.61, 83.15, 82.84, 83.99, 84.55, 84.36, 85.53, 86.54, 86.89, 87.77, 87.29]);
const volume = Float64Array.from([5653100, 6447400, 7690900, 3831400, 4455100, 3798000, 3936200, 4732000, 4841300, 3915300, 6830800, 6694100, 5293600, 7985800, 4807900]);
const [outputs, state] = ti.nvi.indicator([close, volume], []);
console.log('NVI:', outputs[0]);
// State continuation
const n = close.length - 5;
const [, state2] = ti.nvi.indicator([close.slice(0, n), volume.slice(0, n)], []);
const continued = state2.batchIndicator([close.slice(n), volume.slice(n)]);
console.log('Continued NVI:', continued[0]);
import { init } from 'tulip-rs-wasm';
import * as ti from 'tulip-rs-wasm';
await init(); // bundler resolves the WASM asset automatically
const close = [81.59, 81.06, 82.87, 83.00, 83.61, 83.15, 82.84, 83.99, 84.55, 84.36, 85.53, 86.54, 86.89, 87.77, 87.29];
const volume = [5653100, 6447400, 7690900, 3831400, 4455100, 3798000, 3936200, 4732000, 4841300, 3915300, 6830800, 6694100, 5293600, 7985800, 4807900];
const [outputs, state] = ti.nvi.indicator([close, volume], []);
console.log('NVI:', outputs[0]);
// State continuation
const n = close.length - 5;
const [, state2] = ti.nvi.indicator([close.slice(0, n), volume.slice(0, n)], []);
const continued = state2.batchIndicator([close.slice(n), volume.slice(n)]);
console.log('Continued NVI:', continued[0]);
SIMD¶
By assets — same options, N assets in parallel:
use tulip_rs::indicators::nvi::{Nvi, Indicator};
let inputs: [&[&[f64]; 2]; 4] = [
&[c1.as_slice(), v1.as_slice()],
&[c2.as_slice(), v2.as_slice()],
&[c3.as_slice(), v3.as_slice()],
&[c4.as_slice(), v4.as_slice()],
];
let results = Nvi::indicator_by_assets::<4>(&inputs, &[], None).unwrap();
for (i, asset_outputs) in results.iter().enumerate() {
println!("Asset {}: {:?}", i + 1, asset_outputs[0]);
}
This indicator has no options, so by-options SIMD does not apply.
By assets — same option applied to 4 assets in one call (N must be 2/4/8/16):
const double c1[] = {81.59, 81.06, 82.87, 83.00, 83.61, 83.15, 82.84, 83.99, 84.55, 84.36};
const double v1[] = {5653100.0, 6447400.0, 7690900.0, 3831400.0, 4455100.0, 3798000.0,
3936200.0, 4732000.0, 4841300.0, 3915300.0};
const double c2[] = {85.00, 85.50, 86.00, 86.50, 87.00, 87.50, 88.00, 88.50, 89.00, 89.50};
const double v2[] = {4500000.0, 5500000.0, 6500000.0, 4000000.0, 4500000.0, 3800000.0,
4200000.0, 5000000.0, 5100000.0, 4100000.0};
const double c3[] = {78.00, 79.00, 80.00, 81.00, 82.00, 83.00, 84.00, 85.00, 86.00, 87.00};
const double v3[] = {3500000.0, 4500000.0, 5500000.0, 3000000.0, 3500000.0, 2800000.0,
3200000.0, 4000000.0, 4100000.0, 3100000.0};
const double c4[] = {95.00, 96.00, 97.00, 98.00, 99.00, 100.00, 101.00, 102.00, 103.00, 104.00};
const double v4[] = {6500000.0, 7500000.0, 8500000.0, 6000000.0, 6500000.0, 5800000.0,
6200000.0, 7000000.0, 7100000.0, 6100000.0};
const double *const asset1[NVI_INPUTS] = {c1, v1};
const double *const asset2[NVI_INPUTS] = {c2, v2};
const double *const asset3[NVI_INPUTS] = {c3, v3};
const double *const asset4[NVI_INPUTS] = {c4, v4};
const double *const *const simd_inputs[4] = {asset1, asset2, asset3, asset4};
CSimdResult r = nvi_simd_by_assets(simd_inputs, 4, 10, options, NULL, 0);
for (uintptr_t i = 0; i < r.num_results; i++) {
/* r.outputs[i][0] -> asset i's NVI series, length r.output_lens[i][0] */
nvi_state_free(r.states[i]);
}
tulip_ffi_simd_result_free(r);
C FFI offers only by-assets for this indicator (no options).
By assets — same options applied to 4 assets in parallel (lane counts 2/4/8/16):
c1 := []float64{81.59, 81.06, 82.87, 83.00, 83.61, 83.15, 82.84, 83.99, 84.55, 84.36}
v1 := []float64{5653100.0, 6447400.0, 7690900.0, 3831400.0, 4455100.0, 3798000.0,
3936200.0, 4732000.0, 4841300.0, 3915300.0}
// Reuse the same data for assets 2–4 in this example
c2, v2 := c1, v1
c3, v3 := c1, v1
c4, v4 := c1, v1
assets := [][indicators.NviInputs][]float64{
{c1, v1},
{c2, v2},
{c3, v3},
{c4, v4},
}
sim, _ := indicators.Nvi.SimdByAssets(assets, []float64{}, nil)
for i, lanes := range sim.Results {
fmt.Printf("Asset %d: %v\n", i+1, lanes[0])
}
sim.Close() // frees every lane state, then the SIMD buffers
This indicator has no options, so by-options SIMD does not offer (nil options arg).
By assets — same options applied to 4 assets in parallel (N must be 2, 4, 8, or 16):
import org.tuliprs.*;
import org.tuliprs.indicators.Nvi;
double[] c1 = {81.59, 81.06, 82.87, 83.00, 83.61, 83.15, 82.84, 83.99, 84.55, 84.36};
double[] v1 = {5653100.0, 6447400.0, 7690900.0, 3831400.0, 4455100.0, 3798000.0,
3936200.0, 4732000.0, 4841300.0, 3915300.0};
// Reuse the same data for assets 2–4 in this example
double[] c2 = c1;
double[] v2 = v1;
double[] c3 = c1;
double[] v3 = v1;
double[] c4 = c1;
double[] v4 = v1;
// One entry per asset; each asset lists its INPUTS series.
double[][][] assets = {{c1, v1}, {c2, v2}, {c3, v3}, {c4, v4}};
try (SimdResult sim = Nvi.simdByAssets(assets, new double[] {}, null)) {
for (int i = 0; i < sim.numResults(); i++) {
System.out.printf("Asset %d: %s%n", i + 1,
java.util.Arrays.toString(sim.toDoubleArray(i, 0)));
}
} // frees every lane state, then the SIMD buffers (contractual order)
This indicator has no options, so by-options SIMD does not apply.
By assets — same options, N assets in parallel (must be 2, 4, 8, or 16):
simd_inputs = [
[c1, v1],
[c2, v2],
[c3, v3],
[c4, v4],
]
outputs_list, states = tulip_rs.indicators.nvi.simd_by_assets(simd_inputs, [])
for i, asset_outputs in enumerate(outputs_list):
print(f"Asset {i+1}: {asset_outputs[0]}")
This indicator has no options, so by-options SIMD does not apply.
By assets — applied to 4 assets in parallel:
const simdInputs = [
[close.slice(), volume.slice()],
[close.map(v => v * 1.1), volume.map(v => v * 1.1)],
[close.map(v => v * 0.9), volume.map(v => v * 0.9)],
[close.map(v => v * 1.02), volume.map(v => v * 1.02)],
];
const [results] = ti.nvi.simdByAssets(simdInputs, []);
results.forEach((out, i) => console.log(`Asset ${i + 1}:`, out[0]));
This indicator has no options, so by-options SIMD does not apply.