Algorithmic Trading Evaluation Report for Aggressive Stocks – October 2019

Algorithmic Trading Executive Summary

The purpose of this report is to present the results of live algorithmic trading forecast performance evaluation, specifically for Aggressive stocks package provided by I Know First. The following results were observed when signal and predictability filters were applied in order to pick the best performing stocks out of the most predictable ones. The period under evaluation is from 21st October 2018 to 21st October 2019. The corresponding returns distribution by horizon and signal filter level is below:

aggressive stocks algorithmic trading returns short
aggressive stocks algorithmic trading returns long

Aggressive Stocks Package Highlights:

  • The highest algorithmic trading return of 29.33% came from the Top 5 Signals stocks on 1-year investment horizon
  • There is a general increasing trend for returns improvement with the time horizon increase in the Top 5 assets by signal subset
  • Top 5 stocks subset consistently out-perform the S&P 500 benchmark by more than 5 times

Note that the above algorithmic trading results were obtained based on evaluation conducted over the specific time period and using a sample approach of consecutive filtering by predictability and by signal indicators to give a general presentation of the algorithmic trading forecast performance patterns for assets in the Aggressive package. The following report provides extensive explanation on our methodology and detailed analysis of the performance metrics that we obtained during the evaluation. This report continues I Know First evaluation series illustrating the ability to provide successful algorithmic trading predictions for aggressive stocks.

About the I Know First Algorithmic Trading Predictions

tech giants

The I Know First self-learning algorithm analyses, models, and generates the stock market forecast. The algorithm is based on Artificial Intelligence (AI) and Machine Learning (ML), and incorporates elements of Artificial Neural Networks and Genetic Algorithms.

The system outputs the predicted trend as a number, positive or negative, along with a wave chart that predicts how the waves will overlap the trend. This helps the trader to decide which direction to trade, at what point to enter the trade, and when to exit. Since the model is 100% empirical, the results are based only on factual data, thereby avoiding any biases or emotions that may accompany human derived assumptions. The human factor is only involved in building the mathematical framework and providing the initial set of inputs and outputs to the system. The algorithm produces a forecast with a signal and a predictability indicator. The signal is the number in the middle of the box. The predictability is the number at the bottom of the box. At the top, a specific asset is identified. This format is consistent across all predictions.

Our algorithm provides two independent indicators for each asset – Signal and Predictability.

The Signal is the predicted strength and direction of movement of the asset. Measured from -inf to +inf.

The predictability indicates our confidence in that result. It is a Pearson correlation coefficient between past algorithmic performance and actual market movement. Measured from -1 to 1.

You can find the detailed description of our heatmap here.

The Stock Picking Method

The method in this evaluation is as follows:

We take the top 10 assets recommended for long and top 10 recommended for short from each daily forecast. Afterwards, we start filtering union of these subsets by absolute signal value and pick the Y highest signals. Such methodology is employed due to the fact that aggressive stocks covered in this package tend to have significant volatility which makes predictability indicator harder to use and requiring more complex picking strategies, so for the purposes of this specific report that employs simple buy-sell strategy it is omitted from consideration.

For example, a top 10 signal filter means that on each day we take only the 10 assets with the strongest signals by absolute value and pick position for the asset in accordance with the forecast. By that, we essentially utilize a strategy that has long and short positions. If the signal is positive, then we buy and, if negative, we short.

The Performance Evaluation Method

We perform evaluations on the individual forecast level. It means that we calculate what would be the return of each forecast we have issued for each horizon in the testing period. Then, we take the average of those results by strategy and forecast horizon.

For example, to evaluate the performance of our 1-month forecasts, we calculate the return of each trade by using this formula:

This simulates a client purchasing the asset based on our prediction and selling it exactly 1 month in the future.

We iterate this calculation for all trading days in the analyzed period and average the results.

Note that this evaluation does not take a set portfolio and follow it. This is a different evaluation method at the individual forecast level.

The Hit Ratio Calculation

The hit ratio helps us to identify the accuracy of our algorithm’s predictions.

The hit ratio is then calculated as follows:

For instance, a 90% hit ratio for top 10 signal filter would imply that the algorithm correctly predicted the price movements of 9 out of 10 assets within this particular set of assets. If it was a one asset case, as it was in Apple stock forecast evaluation or performed for gold price predictions, the computation would be performed on the number of times that a specific asset was correctly predicted.

The Benchmarking Method

The benchmark used in this report is the average S&P 500 return for the periods of the respective time horizons from 21 October 2018 to 21 October 2019 in a long position. We measured the returns of our package emulating long and short strategy for our forecasts against the benchmark. This helps us to determine the effectiveness of our algorithm by comparing the rate of return of the benchmark with the rate of return of our predictability-based strategy.

Stock Universe Under Consideration – Aggressive Stocks Universe

In this report we conduct algorithmic quantitative trading back-testing for “aggressive” stocks that I Know First cover by its algorithmic forecast in Aggressive stocks package. The period for evaluation and testing is from 21st October 2018 to 21st October 2019. During this period, we were providing our clients with daily forecasts for Aggressive stocks and the time horizons which we evaluate in this report are 6 periods spanning from 3 days to 1 year.

Evaluating the Signal Indicator for Aggressive Stocks

In this section we will demonstrate how adding the signal indicator to our stock picking method improves the above performance even further. After filtering by predictability, we applied further filtering by signal strength to investigate potential improvement. The results of the testing showed that there is a significant positive marginal effect on the assets return, especially in the case of the 14-days, 1-months and 1-year investment horizons. We present our findings in the following table and charts.

aggressive stocks algorithmic trading returns table
Table 1: Package performance in terms of average returns
aggressive stocks algorithmic trading returns short
aggressive stocks algorithmic trading returns long
aggressive stocks hit ratio table
Table 2: Package performance in terms of hit ratios
aggressive stocks hit ratio short
aggressive stocks hit ratio long

From table 1, we can see that if we apply signal strength filtering to the Aggressive stocks’ universe, subsets for 14 days, 1-month and 1-year time horizons will show significant improvements in the performance as we filter from All signals to Top 10, and to Top 5 signals. In general, all the above subsets start to produce greater returns than the benchmark’s for majority of the time horizons, with only exception – 3-months time horizon. As soon as we start to consider longer time horizon, we see that the returns of the Top 10 and Top 5 subsets make significant jumps on this horizons reaching as high as 5.64% and 5.62% by Top 10 and Top 5 on 1-month horizon, respectively; while for 1-year horizon these figures tops at 21.09% and 29.33%, respectively. From table 2, it is easy to see that these high returns come from relatively risky set of stocks, as the hit ratio ranges from 50% to 60%, which is expected as these stocks are characterized by extreme volatility and are extremely complex to forecast. Despite everything, the package out-performs S&P 500 benchmark multiple times, providing an investor with an invaluable tool to capture the most profitable trades in the market.

Conclusions for Algorithmic Trading

In this analysis, we demonstrated the out-performance of our forecasts for the stocks from Aggressive stocks package picked by I Know First’s AI Algorithm for the period from 21st October 2019 to 21st October 2019. 

Applying our predictability indicator as an investment criterion coupled with filtering by our signal strength, results in even greater performance over the benchmarks comprised of stocks from the Aggressive stocks universe. That said, the Top 5 stocks by signal yield significantly higher returns than any other asset subset on all considered time horizons spanning from 3 days to 1 year. Therefore, an investor who wants to critically contribute to the structure of his investments by adding some aggressive stocks to his portfolio can do so by simultaneously utilizing the I Know First predictability and signal indicators as criteria for picking best stocks.