Application of Machine Learning Technology in Trading (3)

fmz.com (34)in #fmz • 9 months ago

Linear regression with normalization

Mean squared error: 0.05
Variance score: 0.90

This model does not improve the previous model, but it is not worse. Now we can compare the coefficients to see which ones are important actually.

Let's look at the coefficients:

for i in range(len(basis_X_train.columns)):
    print('%.4f, %s'%(regr_norm.coef_[i], basis_X_train.columns[i]))

The results are:

19.8727, emabasis4
-9.2015, emabasis5
8.8981, emabasis7
-5.5692, emabasis10
-0.0036, rsi15
-0.0146, rsi10
0.0196, mom10
-0.0035, mom5
-7.9138, basis
0.0062, swidth
0.0117, fwidth
2.0883, btopask
2.0311, btopbid
0.0974, bavgask
0.0611, bavgbid
0.0007, topaskvolratio
0.0113, topbidvolratio
-0.0220, totalaskvolratio
0.0231, totalbidvolratio

We can clearly see that some features have higher coefficients than others, and they may have stronger prediction ability.

Let's see the correlation between different features.

import seaborn

c = basis_X_train.corr()
plt.figure(figsize=(10,10))
seaborn.heatmap(c, cmap='RdYlGn_r', mask = (np.abs(c) <= 0.8))
plt.show()

Correlation between features

The dark red areas represent highly correlated variables. Let's create/modify some features again and try to improve our model.

For example, I can discard features like emabasisdi7 easily, which are only linear combinations of other features.

def create_features_again(data):
    basis_X = pd.DataFrame(index = data.index, columns =  [])
    basis_X['mom10'] = difference(data['basis'],11)
    basis_X['emabasis2'] = ewm(data['basis'],2)
    basis_X['emabasis5'] = ewm(data['basis'],5)
    basis_X['emabasis10'] = ewm(data['basis'],10)
    basis_X['basis'] = data['basis']
    basis_X['totalaskvolratio'] = (data['stockTotalAskVol']
                                 - data['futureTotalAskVol'])/
                                   100000
    basis_X['totalbidvolratio'] = (data['stockTotalBidVol']
                                 - data['futureTotalBidVol'])/
                                   100000
    basis_X = basis_X.fillna(0)
    
    basis_y = data['Y(Target)']
    basis_y.dropna(inplace=True)
    return basis_X, basis_y
basis_X_test, basis_y_test = create_features_again(validation_data)
basis_X_train, basis_y_train = create_features_again(training_data)
_, basis_y_pred = linear_regression(basis_X_train, basis_y_train, basis_X_test,basis_y_test)
basis_y_regr = basis_y_pred.copy()

('Coefficients: ', array([ 0.03246139,
0.49780982, -0.22367172,  0.20275786,  0.50758852,
-0.21510795, 0.17153884]))
Mean squared error: 0.02
Variance score: 0.96

Look, the performance of our model has not changed. We only need some characteristics to explain our target variables. I suggest you try more of the above features, try new combinations, etc., to see what can improve our model.

We can also try more complex models to see whether changes in models can improve performance.

K-Nearest Neighbor (KNN) algorithm

from sklearn import neighbors
n_neighbors = 5
model = neighbors.KNeighborsRegressor(n_neighbors, weights='distance')
model.fit(basis_X_train, basis_y_train)
basis_y_pred = model.predict(basis_X_test)
basis_y_knn = basis_y_pred.copy()

from sklearn.svm import SVR
model = SVR(kernel='rbf', C=1e3, gamma=0.1)
model.fit(basis_X_train, basis_y_train)
basis_y_pred = model.predict(basis_X_test)
basis_y_svr = basis_y_pred.copy()

Decision tree

model=ensemble.ExtraTreesRegressor()
model.fit(basis_X_train, basis_y_train)
basis_y_pred = model.predict(basis_X_test)
basis_y_trees = basis_y_pred.copy()

Step 7: Backtest the test data

Check the performance of the actual sample data

Backtesting performance on (untouched) test data sets

This is a critical moment. We run our final optimization model from the last step of the test data, we put it aside at the beginning and we have not touched the data so far.

This gives you a realistic expectation of how your model will execute on new and unseen data when you start real-time trading. Therefore, it is necessary to ensure that you have a clean data set that is not used to train or verify the model.

If you do not like the backtest results of the test data, please discard the model and start again. Never go back or re-optimize your model, which will lead to over fitting! (It is also recommended to create a new test data set, because this data set is now polluted; when discarding the model, we already know the content of the data set implicitly).

Here we will still use Auquan's Toolbox:

import backtester
from backtester.features.feature import Feature
from backtester.trading_system import TradingSystem
from backtester.sample_scripts.fair_value_params import FairValueTradingParams
class Problem1Solver():
def getTrainingDataSet(self):
        return "trainingData1"
def getSymbolsToTrade(self):
        return ['MQK']
def getCustomFeatures(self):
        return {'my_custom_feature': MyCustomFeature}
def getFeatureConfigDicts(self):
                            
        expma5dic = {'featureKey': 'emabasis5',
                 'featureId': 'exponential_moving_average',
                 'params': {'period': 5,
                              'featureName': 'basis'}}
        expma10dic = {'featureKey': 'emabasis10',
                 'featureId': 'exponential_moving_average',
                 'params': {'period': 10,
                              'featureName': 'basis'}}                     
        expma2dic = {'featureKey': 'emabasis3',
                 'featureId': 'exponential_moving_average',
                 'params': {'period': 3,
                              'featureName': 'basis'}}
        mom10dic = {'featureKey': 'mom10',
                 'featureId': 'difference',
                 'params': {'period': 11,
                              'featureName': 'basis'}}
        
        return [expma5dic,expma2dic,expma10dic,mom10dic]    
    
    def getFairValue(self, updateNum, time, instrumentManager):
        # holder for all the instrument features
        lbInstF = instrumentManager.getlookbackInstrumentFeatures()
        mom10 = lbInstF.getFeatureDf('mom10').iloc[-1]
        emabasis2 = lbInstF.getFeatureDf('emabasis2').iloc[-1]
        emabasis5 = lbInstF.getFeatureDf('emabasis5').iloc[-1]
        emabasis10 = lbInstF.getFeatureDf('emabasis10').iloc[-1] 
        basis = lbInstF.getFeatureDf('basis').iloc[-1]
        totalaskvol = lbInstF.getFeatureDf('stockTotalAskVol').iloc[-1] - lbInstF.getFeatureDf('futureTotalAskVol').iloc[-1]
        totalbidvol = lbInstF.getFeatureDf('stockTotalBidVol').iloc[-1] - lbInstF.getFeatureDf('futureTotalBidVol').iloc[-1]
        
        coeff = [ 0.03249183, 0.49675487, -0.22289464, 0.2025182, 0.5080227, -0.21557005, 0.17128488]
        newdf['MQK'] = coeff[0] * mom10['MQK'] + coeff[1] * emabasis2['MQK'] +\
                      coeff[2] * emabasis5['MQK'] + coeff[3] * emabasis10['MQK'] +\
                      coeff[4] * basis['MQK'] + coeff[5] * totalaskvol['MQK']+\
                      coeff[6] * totalbidvol['MQK']
                    
        newdf.fillna(emabasis5,inplace=True)
        return newdf
problem1Solver = Problem1Solver()
tsParams = FairValueTradingParams(problem1Solver)
tradingSystem = TradingSystem(tsParams)
tradingSystem.startTrading(onlyAnalyze=False, 
                           shouldPlot=True,
                           makeInstrumentCsvs=False)