|
3 registered members (3run, flink, henrybane),
898
guests, and 7
spiders. |
|
Key:
Admin,
Global Mod,
Mod
|
|
|
Multi-Factor Gaussian FX Strategy
[Re: TipmyPip]
#488292
07/06/24 15:56
07/06/24 15:56
|
Joined: Sep 2017
Posts: 93
TipmyPip
OP
Junior Member
|
OP
Junior Member
Joined: Sep 2017
Posts: 93
|
Following more factors of correlation, and Market sentiments, we can also include factor exposure, market capitalization, and industry data from various data sources on the web : #include <default.c>
var alpha, beta;
var filt, filttr, hband, lband;
vars PriceClose, GaussianFiltered, GaussianFilteredTR;
// Define additional factor variables
vars FactorExposure, MarketCap, IndustryData;
var beta0, betaFactor, betaMktCap, betaIndustry;
// Initialize factors for each time period
void initFactors() {
FactorExposure = series(0);
MarketCap = series(0);
IndustryData = series(0);
// Set coefficients based on your model or optimization
beta0 = 0.01;
betaFactor = 0.02;
betaMktCap = 0.03;
betaIndustry = 0.04;
}
void init() {
BarPeriod = 60; // 1-hour bars
LookBack = 144; // Lookback period for the Gaussian filter
PriceClose = series(priceClose());
GaussianFiltered = series(0);
GaussianFilteredTR = series(0);
alpha = calculateAlpha(144, 4); // Sampling Period and Poles
beta = (1 - cos(2 * PI / 144)) / (pow(1.414, 2 / 4) - 1);
initFactors();
}
var calculateAlpha(int period, int poles) {
var beta = (1 - cos(2 * PI / period)) / (pow(1.414, 2 / poles) - 1);
return -beta + sqrt(beta * beta + 2 * beta);
}
void calculateGaussianFilter(vars src, int poles, int period) {
vars filtSeries = series(0);
vars filtSeriesTR = series(0);
var lag = (period - 1) / (2 * poles);
vars srcData = series(0);
vars trData = series(0);
for(int i = 0; i < Bar; i++) {
if(i > lag) {
srcData[i] = src[i] + (src[i] - src[i - lag]);
trData[i] = TR() + (TR() - TR(lag));
} else {
srcData[i] = src[i];
trData[i] = TR();
}
}
for(int i = 0; i < Bar; i++) {
filtSeries[i] = filt9x(alpha, srcData[i], 1, poles);
filtSeriesTR[i] = filt9x(alpha, trData[i], 1, poles);
}
GaussianFiltered[0] = filtSeries[0];
GaussianFilteredTR[0] = filtSeriesTR[0];
hband = GaussianFiltered[0] + GaussianFilteredTR[0] * 1.414;
lband = GaussianFiltered[0] - GaussianFilteredTR[0] * 1.414;
}
var filt9x(var a, var s, int i, int poles) {
static vars f[10];
var x = 1 - a;
var filt = pow(a, i) * s + i * x * f[1] - (i >= 2 ? 36 * pow(x, 2) * f[2] : 0) + (i >= 3 ? 84 * pow(x, 3) * f[3] : 0)
- (i >= 4 ? 126 * pow(x, 4) * f[4] : 0) + (i >= 5 ? 126 * pow(x, 5) * f[5] : 0) - (i >= 6 ? 84 * pow(x, 6) * f[6] : 0)
+ (i >= 7 ? 36 * pow(x, 7) * f[7] : 0) - (i >= 8 ? 9 * pow(x, 8) * f[8] : 0) + (i == 9 ? 1 * pow(x, 9) * f[9] : 0);
for(int j = 9; j > 0; j--) {
f[j] = f[j - 1];
}
f[0] = filt;
return filt;
}
bool crossOver(vars data, var level) {
return (data[1] < level && data[0] >= level);
}
bool crossUnder(vars data, var level) {
return (data[1] > level && data[0] <= level);
}
// Predict returns based on factors
var predictReturn() {
var predictedReturn = beta0 + betaFactor * FactorExposure[0] + betaMktCap * MarketCap[0] + betaIndustry * IndustryData[0];
return predictedReturn;
}
void run() {
if(is(INITRUN)) {
init();
}
calculateGaussianFilter(PriceClose, 4, 144);
bool longCondition = crossOver(priceClose(), hband) && predictReturn() > 0;
bool closeAllCondition = crossUnder(priceClose(), hband) || predictReturn() < 0;
if (longCondition) {
enterLong();
}
if (closeAllCondition) {
exitLong();
}
}
Last edited by TipmyPip; 07/06/24 15:58.
|
|
|
Gaussian Bands Strategy
[Re: TipmyPip]
#488293
07/06/24 17:16
07/06/24 17:16
|
Joined: Sep 2017
Posts: 93
TipmyPip
OP
Junior Member
|
OP
Junior Member
Joined: Sep 2017
Posts: 93
|
Here is another version for the Gaussian Filter Strategy, while it can be considered non-linear under multi-correlation factors of overlapping data : #include <default.c>
// Variables
var alpha, beta;
var filt, filttr, hband, lband;
vars PriceClose, GaussianFiltered, GaussianFilteredTR;
vars UpperBand, LowerBand;
// Initialization
void init() {
BarPeriod = 60; // 1-hour bars
LookBack = 144; // Lookback period for the Gaussian filter
PriceClose = series(priceClose());
GaussianFiltered = series(0);
GaussianFilteredTR = series(0);
UpperBand = series(0);
LowerBand = series(0);
alpha = calculateAlpha(144, 4); // Sampling Period and Poles
beta = (1 - cos(2 * PI / 144)) / (pow(1.414, 2 / 4) - 1);
}
// Calculate Alpha
var calculateAlpha(int period, int poles) {
var beta = (1 - cos(2 * PI / period)) / (pow(1.414, 2 / poles) - 1);
return -beta + sqrt(beta * beta + 2 * beta);
}
// Modified Gaussian Filter with Bands
var filt9x_with_bands(var a, var s, int poles, var deviationMultiplier, var* upperBand, var* lowerBand) {
static vars f[10]; // Array to store previous filter values
var x = 1 - a;
var filt = s * a;
// Precompute powers of x to avoid redundant calculations
var x2 = x * x;
var x3 = x2 * x;
var x4 = x3 * x;
var x5 = x4 * x;
var x6 = x5 * x;
var x7 = x6 * x;
var x8 = x7 * x;
var x9 = x8 * x;
// Calculate the filter value iteratively
for (int i = 1; i <= poles; i++) {
switch (i) {
case 1:
filt += x * f[i];
break;
case 2:
filt -= 36 * x2 * f[i];
break;
case 3:
filt += 84 * x3 * f[i];
break;
case 4:
filt -= 126 * x4 * f[i];
break;
case 5:
filt += 126 * x5 * f[i];
break;
case 6:
filt -= 84 * x6 * f[i];
break;
case 7:
filt += 36 * x7 * f[i];
break;
case 8:
filt -= 9 * x8 * f[i];
break;
case 9:
filt += x9 * f[i];
break;
}
}
// Shift the previous values in the array
for (int j = 9; j > 0; j--) {
f[j] = f[j - 1];
}
f[0] = filt;
// Calculate standard deviation of filter values
var sum = 0, mean = 0, count = 10;
for (int j = 0; j < count; j++) {
sum += f[j];
}
mean = sum / count;
var varianceSum = 0;
for (int j = 0; j < count; j++) {
varianceSum += (f[j] - mean) * (f[j] - mean);
}
var stddev = sqrt(varianceSum / count);
// Calculate upper and lower bands
*upperBand = filt + deviationMultiplier * stddev;
*lowerBand = filt - deviationMultiplier * stddev;
return filt;
}
// Gaussian Filter Calculation
void calculateGaussianFilter(vars src, int poles, int period) {
vars filtSeries = series(0);
vars filtSeriesTR = series(0);
var lag = (period - 1) / (2 * poles);
vars srcData = series(0);
vars trData = series(0);
for(int i = 0; i < Bar; i++) {
if(i > lag) {
srcData[i] = src[i] + (src[i] - src[i - lag]);
trData[i] = TR() + (TR() - TR(lag));
} else {
srcData[i] = src[i];
trData[i] = TR();
}
}
for(int i = 0; i < Bar; i++) {
var upper, lower;
filtSeries[i] = filt9x_with_bands(alpha, srcData[i], poles, 1.414, &upper, &lower);
filtSeriesTR[i] = filt9x_with_bands(alpha, trData[i], poles, 1.414, &upper, &lower);
UpperBand[i] = upper;
LowerBand[i] = lower;
}
GaussianFiltered[0] = filtSeries[0];
GaussianFilteredTR[0] = filtSeriesTR[0];
hband = UpperBand[0];
lband = LowerBand[0];
}
// CrossOver and CrossUnder Functions
bool crossOver(vars data, var level) {
return (data[1] < level && data[0] >= level);
}
bool crossUnder(vars data, var level) {
return (data[1] > level && data[0] <= level);
}
// Main Trading Logic
void run() {
if(is(INITRUN)) {
init();
}
calculateGaussianFilter(PriceClose, 4, 144);
bool longCondition = crossOver(priceClose(), hband);
bool closeAllCondition = crossUnder(priceClose(), lband);
if (longCondition) {
enterLong();
}
if (closeAllCondition) {
exitLong();
}
}
|
|
|
Gaussian Decision Tree Hedging Strategy
[Re: TipmyPip]
#488305
07/11/24 05:43
07/11/24 05:43
|
Joined: Sep 2017
Posts: 93
TipmyPip
OP
Junior Member
|
OP
Junior Member
Joined: Sep 2017
Posts: 93
|
This trading strategy leverages advanced signal processing and machine learning techniques to make informed trading decisions. We employ a Gaussian filter, which smooths the price series by reducing noise, and derive key indicators from it. These indicators include the filtered price and its bands, calculated by a non-recursive method. The signals are generated by comparing the filtered price to its bands, and identifying potential entry and exit points. To enhance decision-making, we integrate a decision tree algorithm. This machine learning model is trained on historical signals, capturing complex patterns and relationships in the data. The model predicts future price movements, guiding long and short positions. By combining signal processing with machine learning, the strategy aims to exploit market inefficiencies and improve trading performance. #include <default.c>
// Variables
var alpha, beta;
var filt, filttr, hband, lband;
vars PriceClose, GaussianFiltered, GaussianFilteredTR;
vars UpperBand, LowerBand;
// Initialization
void init() {
BarPeriod = 60; // 1-hour bars
LookBack = 150; // Lookback period for the Gaussian filter
PriceClose = series(priceClose());
GaussianFiltered = series(0);
GaussianFilteredTR = series(0);
UpperBand = series(0);
LowerBand = series(0);
alpha = calculateAlpha(144, 4); // Sampling Period and Poles
beta = (1 - cos(2 * PI / 144)) / (pow(1.414, 2 / 4) - 1);
adviseLong(DTREE, 0, NULL, 0); // Initialize the Decision Tree model
}
// Calculate Alpha
var calculateAlpha(int period, int poles) {
var beta = (1 - cos(2 * PI / period)) / (pow(1.414, 2 / poles) - 1);
return -beta + sqrt(beta * beta + 2 * beta);
}
// Modified Gaussian Filter with Bands
var filt9x_with_bands(var a, var s, int poles, var deviationMultiplier, var* upperBand, var* lowerBand) {
static vars f[10]; // Array to store previous filter values
var x = 1 - a;
var filt = s * a;
// Precompute powers of x to avoid redundant calculations
var x2 = x * x;
var x3 = x2 * x;
var x4 = x3 * x;
var x5 = x4 * x;
var x6 = x5 * x;
var x7 = x6 * x;
var x8 = x7 * x;
var x9 = x8 * x;
// Calculate the filter value iteratively
for (int i = 1; i <= poles; i++) {
switch (i) {
case 1:
filt += x * f[i];
break;
case 2:
filt -= 36 * x2 * f[i];
break;
case 3:
filt += 84 * x3 * f[i];
break;
case 4:
filt -= 126 * x4 * f[i];
break;
case 5:
filt += 126 * x5 * f[i];
break;
case 6:
filt -= 84 * x6 * f[i];
break;
case 7:
filt += 36 * x7 * f[i];
break;
case 8:
filt -= 9 * x8 * f[i];
break;
case 9:
filt += x9 * f[i];
break;
}
}
// Shift the previous values in the array
for (int j = 9; j > 0; j--) {
f[j] = f[j - 1];
}
f[0] = filt;
// Calculate standard deviation of filter values
var sum = 0, mean = 0, count = 10;
for (int j = 0; j < count; j++) {
sum += f[j];
}
mean = sum / count;
var varianceSum = 0;
for (int j = 0; j < count; j++) {
varianceSum += (f[j] - mean) * (f[j] - mean);
}
var stddev = sqrt(varianceSum / count);
// Calculate upper and lower bands
*upperBand = filt + deviationMultiplier * stddev;
*lowerBand = filt - deviationMultiplier * stddev;
return filt;
}
// Gaussian Filter Calculation
void calculateGaussianFilter(vars src, int poles, int period) {
vars filtSeries = series(0);
vars filtSeriesTR = series(0);
var lag = (period - 1) / (2 * poles);
vars srcData = series(0);
vars trData = series(0);
for (int i = 0; i < Bar; i++) {
if (i > lag) {
srcData[i] = src[i] + (src[i] - src[i - lag]);
trData[i] = TR() + (TR() - TR(lag));
} else {
srcData[i] = src[i];
trData[i] = TR();
}
}
for (int i = 0; i < Bar; i++) {
var upper, lower;
filtSeries[i] = filt9x_with_bands(alpha, srcData[i], poles, 1.414, &upper, &lower);
filtSeriesTR[i] = filt9x_with_bands(alpha, trData[i], poles, 1.414, &upper, &lower);
UpperBand[i] = upper;
LowerBand[i] = lower;
}
GaussianFiltered[0] = filtSeries[0];
GaussianFilteredTR[0] = filtSeriesTR[0];
hband = UpperBand[0];
lband = LowerBand[0];
}
// Main Trading Logic
void run() {
BarPeriod = 60; // 1-hour bars
LookBack = 150; // Lookback period
TradesPerBar = 2;
if (Train) {
Hedge = 2; // Hedge during training
}
set(RULES | TESTNOW); // Set rules and test mode
if (is(INITRUN)) {
init();
}
calculateGaussianFilter(PriceClose, 4, 144);
// Generate some signals from GaussianFiltered in the -100..100 range
var Signals[2];
Signals[0] = (GaussianFiltered[0] - GaussianFilteredTR[0]) / PIP;
Signals[1] = 100 * FisherN(priceClose(), 100);
// Train and trade the signals using the decision tree model
if (adviseLong(DTREE, 0, Signals, 2) > 0)
enterLong();
if (adviseShort(DTREE, 0, Signals, 2) > 0)
enterShort();
}
Last edited by TipmyPip; 07/11/24 05:45.
|
|
|
Gaussian-Enhanced Hybrid Ensemble Strategy
[Re: TipmyPip]
#488318
07/20/24 05:00
07/20/24 05:00
|
Joined: Sep 2017
Posts: 93
TipmyPip
OP
Junior Member
|
OP
Junior Member
Joined: Sep 2017
Posts: 93
|
The following is an optimized version of the Gaussian Filter Strategy, which includes Decision Trees, and Neural Networks to monitor error correction within the signals. This code combines decision trees and neural networks to enhance previous versions of the implementation. It initializes and trains both models, uses Gaussian filters for signal processing, and employs a meta-model neural network to integrate their predictions for final trading actions, aiming to reduce errors and improve trading performance in a systematic, data-driven manner. #include <default.c>
// Variables
var alpha, beta;
var filt, filttr, hband, lband;
vars PriceClose, GaussianFiltered, GaussianFilteredTR;
vars UpperBand, LowerBand;
var Signals[2];
var MetaSignals[2]; // Signals for the meta-model
// Initialization
void init() {
BarPeriod = 60; // 1-hour bars
LookBack = 150; // Lookback period for the Gaussian filter
PriceClose = series(priceClose());
GaussianFiltered = series(0);
GaussianFilteredTR = series(0);
UpperBand = series(0);
LowerBand = series(0);
alpha = calculateAlpha(144, 4); // Sampling Period and Poles
beta = (1 - cos(2 * PI / 144)) / (pow(1.414, 2 / 4) - 1);
adviseLong(DTREE, 0, NULL, 0); // Initialize the Decision Tree model
adviseLong(NN, 0, NULL, 0); // Initialize the Neural Network model
adviseLong(NN, 1, NULL, 0); // Initialize the Meta-Model Neural Network
}
// Calculate Alpha
var calculateAlpha(int period, int poles) {
var beta = (1 - cos(2 * PI / period)) / (pow(1.414, 2 / poles) - 1);
return -beta + sqrt(beta * beta + 2 * beta);
}
// Optimized Gaussian Filter with Bands
var filt9x_with_bands(var a, var s, int poles, var deviationMultiplier, var* upperBand, var* lowerBand) {
static vars f[10]; // Array to store previous filter values
var x = 1 - a;
var filt = s * a;
// Precompute powers of x to avoid redundant calculations
var x_powers[10];
x_powers[1] = x;
for (int i = 2; i <= 9; i++) {
x_powers[i] = x_powers[i - 1] * x;
}
// Calculate the filter value iteratively
var coefficients[10] = {0, 1, -36, 84, -126, 126, -84, 36, -9, 1};
for (int i = 1; i <= poles; i++) {
filt += coefficients[i] * x_powers[i] * f[i];
}
// Shift the previous values in the array
for (int j = 9; j > 0; j--) {
f[j] = f[j - 1];
}
f[0] = filt;
// Calculate mean and variance in a single pass
var sum = 0, varianceSum = 0, count = 10;
for (int j = 0; j < count; j++) {
sum += f[j];
}
var mean = sum / count;
for (int j = 0; j < count; j++) {
varianceSum += (f[j] - mean) * (f[j] - mean);
}
var stddev = sqrt(varianceSum / count);
// Calculate upper and lower bands
*upperBand = filt + deviationMultiplier * stddev;
*lowerBand = filt - deviationMultiplier * stddev;
return filt;
}
// Gaussian Filter Calculation
void calculateGaussianFilter(vars src, int poles, int period) {
vars filtSeries = series(0);
vars filtSeriesTR = series(0);
var lag = (period - 1) / (2 * poles);
vars srcData = series(0);
vars trData = series(0);
for (int i = 0; i < Bar; i++) {
if (i > lag) {
srcData[i] = src[i] + (src[i] - src[i - lag]);
trData[i] = TR() + (TR() - TR(lag));
} else {
srcData[i] = src[i];
trData[i] = TR();
}
}
for (int i = 0; i < Bar; i++) {
var upper, lower;
filtSeries[i] = filt9x_with_bands(alpha, srcData[i], poles, 1.414, &upper, &lower);
filtSeriesTR[i] = filt9x_with_bands(alpha, trData[i], poles, 1.414, &upper, &lower);
UpperBand[i] = upper;
LowerBand[i] = lower;
}
GaussianFiltered[0] = filtSeries[0];
GaussianFilteredTR[0] = filtSeriesTR[0];
hband = UpperBand[0];
lband = LowerBand[0];
}
// Meta-Model for Combining Predictions using Neural Network
var metaModel(var dtreePrediction, var nnPrediction) {
MetaSignals[0] = dtreePrediction;
MetaSignals[1] = nnPrediction;
return advise(NN, 1, MetaSignals, 2); // Use another neural network (NN) model as meta-model
}
// Main Trading Logic
void run() {
BarPeriod = 60; // 1-hour bars
LookBack = 150; // Lookback period
TradesPerBar = 2;
if (Train) {
Hedge = 2; // Hedge during training
}
set(RULES | TESTNOW); // Set rules and test mode
if (is(INITRUN)) {
init();
}
calculateGaussianFilter(PriceClose, 4, 144);
// Generate some signals from GaussianFiltered in the -100..100 range
Signals[0] = (GaussianFiltered[0] - GaussianFilteredTR[0]) / PIP;
Signals[1] = 100 * FisherN(priceClose(), 100);
// Train and trade the signals using decision tree and neural network models
var dtreePrediction = advise(DTREE, 0, Signals, 2);
var nnPrediction = advise(NN, 0, Signals, 2);
var finalDecision = metaModel(dtreePrediction, nnPrediction);
if (finalDecision > 0)
enterLong();
else if (finalDecision < 0)
enterShort();
else {
if (isLong()) exitLong();
if (isShort()) exitShort();
}
}
|
|
|
Re: Gaussian-Enhanced Hybrid Ensemble Strategy
[Re: TipmyPip]
#488319
07/20/24 06:46
07/20/24 06:46
|
Joined: Sep 2017
Posts: 93
TipmyPip
OP
Junior Member
|
OP
Junior Member
Joined: Sep 2017
Posts: 93
|
I must confess, my fascination with automation knows no bounds. While resolving coding problems can be a daunting task, the correct implementation of automation can significantly simplify much of this work.
Imagine the incredible efficiency we could achieve by quickly resolving problems and eliminating the need for error corrections in our code. This would not only save us time but also enrich our lives by allowing us to focus on more creative and fulfilling pursuits.
It's true that addressing errors in automated code generation can be a slow process. However, developing and debugging code manually without automation demands just as much effort and time.
My enthusiasm for automation is matched by my eagerness to share its benefits with others. Please forgive my insistence, but I wholeheartedly encourage you to embrace automation as I have. Together, we can transform our manual trading tasks into streamlined, automated solutions. :-) Hahahah!
I surely, will need to many errors, in my code, But if anyone, feels like sharing any solutions, or ideas, please do.
Last edited by TipmyPip; 07/22/24 19:04.
|
|
|
|
