## Executive Summary

In this forecast evaluation report, we will examine the performance of the forecasts generated by the I Know First AI Algorithm for Euro to US dollar with time horizons ranging from 3 days to 1 year, which were delivered daily to all clients. Our analysis covers the time period from 1 January 2018 to 1 July 2019. Below, we present our key takeaways:

### Package Highlights:

• I Know First forecasts for the currency pair of EUR/USD outperformed the benchmark for all time horizons.
• The best performance came from the 1 year forecast, which produced 1.48% return against the benchmark’s -5.75%.

Note that the above results were obtained as a result of evaluation conducted over the specific time period from 1 January 2018 to 1 July 2019 to give a general presentation of the forecasts performance pattern for specific currency pair. The following report provides extensive explanation on our methodology and detailed analysis of the performance metrics that we obtained during the evaluation.

## About the I Know First Algorithm

The I Know First self-learning algorithm analyses, models, and predicts the capital market, including stocks, bonds, currencies, commodities and interest rates. 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 with the predicted trend. Consequently, the trader can decide which direction to trade, when to enter the trade, and when to exit the trade. The model is 100% empirical, based only on factual data, thereby avoiding any biases or emotions that may accompany human assumptions. I Know First’s model only involves the human factor is 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 asset – signal and predictability.

The signal is the predicted strength and direction of movement of the asset. This is measured from -inf to +inf.

The predictability indicates our confidence in the signal. The predictability is a Pearson correlation coefficient relating past algorithmic performance and actual market movement, measured from -1 to 1.

You can find a detailed description of our heatmap here.

## The Performance Evaluation Method

We perform evaluations on the individual forecast level. This means that we calculate the return of each forecast we have issued for each horizon in the testing period. We then take the average of those results based on our positions on different currencies 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 investing into EUR/USD currency pair on the day we issue our prediction and closing it exactly 1 month in the future from that day.

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

## The Hit Ratio Calculation

The hit ratio helps us to identify the accuracy of our algorithm’s predictions. The hit ratio is calculated for each of the forecast horizons as follows:

$\dpi{150}&space;\tiny&space;Hit\&space;Ratio&space;(\%)&space;=&space;\frac{Number\&space;of\&space;instances\&space;EUR/USD\&space;was\&space;predicted\&space;correctly}{Total\&space;number\&space;of\&space;predictions}\times&space;100\%$

For instance, a 90% hit ratio for 3 days horizon would imply that the algorithm correctly predicted the currency pair movements 9 out of 10 times on average for 3 days horizon forecasts during 1 January 2018 to 1 July 2019.

## The Benchmarking Method

The theory behind our benchmarking method is the “Null hypothesis“. The benchmark used in this report is the average of the returns generated by the theoretical long positions taken on Euro to US Dollar pair for the respective time horizons (from 3 days to 1 year) within period from 1 January 2018 to 1 July 2019. We measured the returns of our long and short strategy for our forecasts against this benchmark. By taking the returns difference (referred further as delta) we determine how effective our algorithm in comparison with the benchmark when somebody wants to trade in Euro and US Dollars.

## Performance Evaluation

We conduct our research for the period from 1 January 2018 to 1 July 2019. Following the methodology from the previous sections, we start our analysis by computing the performance of the algorithm’s forecasts for time horizons ranging from 3 days to 1 year. Afterwards, we calculate the returns for the same time horizons for the benchmark using the currencies universe and measured our performance against the benchmark.

## Evaluating the Forecast by Average Returns

In this section, we present our results of the returns generated by our forecasts for EUR/USD against the benchmark.

From the set of charts above one can clearly see that our forecasts generated more returns than the benchmark. The best performance came from the 1 year forecast, which generated return of 1.48%. This is results in delta to the benchmark of 7.23%. While the returns may not be extremely high in terms of percentage, our forecasts managed to avoid making losses for the time horizons of 2 weeks to 1 year, unlike the benchmark which recorded negative performance on all time horizons. Despite facing insignificant losses when 3 days and 1 week forecasts are utilized, our forecasts managed to provide positive and increasing returns for longer forecast horizons, peaking at 1 year with 1.48%.

## Evaluating the Forecast by Out-Performance Delta

It is also important to measure the delta relative to the benchmark. In Figure 3, we can see that our forecast for the currency pair of EUR/USD have provided for positive delta on all time horizons. The best performance came from the 1 year forecast, which featured delta of 7.23%. The general trend to be spotted in Figure 3 is the delta increase by multiples with expansion of the forecast horizon used for trading.

## Evaluating the Forecast by Hit Ratio

Hit ratios are important for the investor who is using I Know First’s AI algorithm. The investor is interested in understanding how his manually composed portfolio would compare against one that is composed using the algorithm. If the hit ratio is 50%, it is merely as good as the flip of a fair coin. If one considers Figure 4, the hit ratios are improving as the investment horizon expands. The peak hit ratio in figure 4 is 75% for the 3 months period.

## Conclusion

In this analysis, we demonstrated the delta out-performance of our forecasts for the currency pair EUR/USD generated by I Know First’s AI Algorithm during the period from 1 January 2018 to 1 July 2019. Based on the presented observations, we record significant delta of the returns from potential trading in EUR/USD currency pair against the theoretical benchmark. As shown in the above diagram, our Long and Short strategy for EUR/USD, which represents mere following the I Know First predictions, yielded significantly higher return on the 1 year time horizon than just holding long positions. Furthermore, we also note that for all time horizons, our forecast for EUR/USD perform better the benchmark.