Executive Summary

In this stock market forecast evaluation report, we will examine the performance of the forecasts generated by the I Know First AI Algorithm for the S&P 500 stocks for long and short positions which were sent daily to our customers. Our analysis covers the period from October 9, 2018, to July 19th, 2020 and further analyzes the performance of the S&P stocks predictions for long and short positions and only long position. 

S&P 500 Stocks Highlights 

• In the short term horizons, all groups’ signals outperformed the S&P 500.
• Our highest signals tend to have the largest returns over short term time horizons and over year long periods. 
• Only long positions had higher returns than the Long and Short for each time horizon.  
• Every signal group across every time horizon gave a hit ratio over 53%, showing consistent and reliable accuracy.
• Our top-5 signals had returns of 12.80% (Long Only) and 11.49% (Long and Short) over year long periods, handily beating the S&P 500 index. 

About I Know First’s Algorithm 

The I Know First self-learning algorithm analyzes, models, and predicts the stock market. 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 the 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 a detailed description of our heatmap here.

Evaluating US Stock Forecasts: S&P 500 Stocks Package

We choose stocks from the S&P 500 by taking the top 30 most predictable assets, and then we apply a set of signal-based filters: top 20, 10 and 5 based on signals. By doing so we focus on the most predictable assets on the one hand, while capturing the ones with the highest signal on the other.

For the analysis we group the forecasts by absolute signals since these strategies are long and short. If the signal is positive, then we buy and if negative, we short. We do not short in the “long only” strategy. 

The Stock Market Forecast 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 Method

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

Using our Daily Forecast asset filtering, we predict the direction of the movement of different assets. Our predictions are then compared against actual movements of these assets within the same time horizon.

The hit ratio is then calculated as follows:

For instance, a 90% hit ratio for a predictability filter with a 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.

The Benchmarking Method: S&P 500 

In order to evaluate our algorithm’s performance, we used the S&P 500 index as a benchmark.

The S&P 500 measures the stock performance of 500 of the U.S’ largest publicly traded companies. It is one of the most followed equity indices and is frequently used as the best gauge of large cap US equities. The S&P is often used as a benchmark for the performance of US publicly traded companies, and the US market as a whole. The S&P 500 is a capitalization-weighted index, the weight of each company in the index is determined based on its market cap divided by the aggregate market cap of all the S&P 500 companies.

For each time horizon, we compare the S&P 500 performance with the performance of our forecasts.

Performance Evaluation: S&P Long Only 

According to table 1, our algorithm outperformed the S&P for all signals between 3-14 days. Our top 5- signals outperformed the benchmark in all time horizons. Over a year-long period, only our top-5 signals beat the S&P 500.

As seen in table 2, all hit ratios had over 50% success. On average, the highest hit ratio was over year long time horizons. The worst hit ratio occurred over 7 day horizons, at 53%. Over longer periods of time and choosing our top signals, we tend to have a higher hit ratio.  

Performance Evaluation: S&P Long and Short 

According to Table 3, from 3-14 days, our algorithm outperformed the S&P 500 for our top-20, top-10, and top-5 stock signals. As the selection of stocks is more concentrated, our top-5 signals performed best, beating the S&P from 3-14 days and having the highest returns yearly. Our top-5 signals had positive returns in all time horizons.

As can be seen by table 4, on average, the year long forecast generated the highest hit ratio. The lowest hit ratio across the board is over 3 month horizons. As we become more selective with our signals, we have a higher hit ratio: Our top 5 signal tend performs better than the other signal groups

Conclusion

This evaluation presented the performance of I Know First’s algorithm for the top S&P 500 Stocks forecast from October 9th, 2018 to July 19th 2020. It shows the average returns and hit ratios from three days to one year, with further evaluation on the algorithms long efficacy. 

I Know First’s algorithm consistently beat the S&P 500’s returns in time horizons less than 14 days. Over a year long time horizon, our highest signals tended to have the largest returns. Our top-5 signals outperformed the S&P 500 constantly. Our algorithm tended to perform better with a focus on long positions. However, both longs and shorts were extremely effective in short term time horizons. It is important to note that every signal group across every time horizon gave a hit ratio over 53%, showing consistent and reliable accuracy. 

It is important to note that this time period included the coronavirus pandemic, including unpredictable times. But, our algorithm has still managed to have good results in the S&P 500. That being said, our algorithm is constantly learning and we look forward to new market data in the following months. We will continue to monitor changes in performance trends and we will communicate to both our investors and subscribers for follow-up reports.