Predicting stocks using the SPX index

This trading strategy is designed for the Quantiacs platform, which hosts competitions for trading algorithms. Detailed information about the competitions is available on the official Quantiacs website.

How to Run the Strategy

In an Online Environment

The strategy can be executed in an online environment using Jupiter or JupiterLab on the Quantiacs personal dashboard. To do this, clone the template in your personal account.

In a Local Environment

To run the strategy locally, you need to install the Quantiacs Toolbox.

Strategy Overview

This file is a Jupyter notebook that contains a detailed implementation for predicting stocks within the NASDAQ 100 index using the SPX index and various technical analysis indicators. The notebook is structured into several markdown and code cells that describe the approach and implement multiple trading strategies. Each strategy utilizes different technical indicators such as the Trix, RSI, and ROC to generate trading signals. The notebook also includes functions to load market data, calculate the indicators, apply trading logic, and evaluate the performance of the strategies by displaying statistics and equity curves. Additionally, there are examples demonstrating the use of moving averages, exponential moving averages, and the Advance–Decline line in trading strategies.

from IPython.display import display
import xarray as xr
import qnt.data as qndata
import qnt.output as qnout
import qnt.ta as qnta
import qnt.stats as qns


def get_SPX(market_data):
    index_name = 'SPX'
    index_data = qndata.index.load_data(assets=[index_name], min_date='2005-01-01', forward_order=True)
    spx_data = index_data.sel(asset=index_name)
    spx_data = xr.align(spx_data, market_data.isel(field=0), join='right')[0]
    return spx_data


def get_strategy_1(data, spx, params):
    def get_trix(prices, index, periods):
        result = prices.copy(True)
        t = qnta.trix(index, periods)
        for a in prices.asset.values:
            result.loc[{"asset": a}] = t
        return result

    trix = get_trix(data.sel(field='close'), spx, 40)

    strategy_1 = trix.shift(time=params[0]) < trix.shift(time=params[1])
    strategy_2 = trix.shift(time=params[2]) > trix.shift(time=params[3])

    weights = strategy_1 * strategy_2 * data.sel(field="is_liquid")
    return weights.fillna(0)


def get_strategy_2(data, spx, params):
    def get_rsi(prices, index, periods):
        result = prices.copy(True)
        r = qnta.rsi(index, periods)
        for a in prices.asset.values:
            result.loc[{"asset": a}] = r
        return result

    rsi = get_rsi(data.sel(field='close'), spx, 40)

    strategy_1 = rsi.shift(time=params[0]) < rsi.shift(time=params[1])
    strategy_2 = rsi.shift(time=params[2]) > rsi.shift(time=params[3])

    weights = strategy_1 * strategy_2 * data.sel(field="is_liquid")
    return weights.fillna(0)


def get_strategy_3(data, spx, params):
    def get_roc(prices, index, periods):
        result = prices.copy(True)
        r = qnta.roc(index, periods)
        for a in prices.asset.values:
            result.loc[{"asset": a}] = r
        return result

    roc = get_roc(data.sel(field='close'), spx, 15)

    strategy_1 = roc.shift(time=params[0]) < roc.shift(time=params[1])
    strategy_2 = roc.shift(time=params[2]) > roc.shift(time=params[3])

    weights = strategy_1 * strategy_2 * data.sel(field="is_liquid")
    return weights.fillna(0)


data = qndata.stocks.load_ndx_data(min_date="2005-01-01")
spx = get_SPX(data)

weights_1_1 = get_strategy_1(data, spx, [142, 54, 132, 63])
weights_1_2 = get_strategy_1(data, spx, [166, 75, 46, 24])
weights_2 = get_strategy_2(data, spx, [159, 78, 77, 167])
weights_3 = get_strategy_3(data, spx, [10, 27, 29, 41])

weights_all = weights_1_1 + weights_1_2 + weights_2 + weights_3
weights = qnout.clean(output=weights_all, data=data, kind="stocks_nasdaq100")


def print_statistic(data, weights_all):
    import plotly.graph_objs as go
    import qnt.stats as qnstats

    stats = qnstats.calc_stat(data, weights_all)
    display(stats.to_pandas().tail(5))

    equity_curve = stats.loc[:, "equity"]
    fig = go.Figure(data=[
        go.Scatter(
            x=equity_curve.time.to_pandas(),
            y=equity_curve,
            hovertext="Equity curve",
        )
    ])
    fig.update_yaxes(fixedrange=False)
    fig.show()


print_statistic(data, weights)
qnout.check(weights, data, "stocks_nasdaq100")
qnout.write(weights)  # To participate in the competition, save this code in a separate cell.