Stock Market Forecast: Does the Stock Market Feature Memory?

This article was written by:

Sergey Okun  Sergey Okun – Financial Analyst at I Know First, Ph.D. in Economics.

Alisa IartsevaAlisa Iartseva – Data analyst at I Know First.


  • The stock market has a long-term memory that allows us to make reliable predictions of the future based on previous behaviors and tendencies.
  • The S&P 500 has the highest Hurst exponent compared with the DAX 30, the CAC 40, and the FTSE 100.
  • The Hurst exponent rises with the extension of the step period length which decreases noise in a time series and increases predictability.

premiumRead The Full Premium Article

Subscribe to receive exclusive PREMIUM content here

World Indices Forecast: AI Beats the S&P500 by 41.85%


  • The highest average return is 57.52% for the Top 5 Signals on a 1-year time horizon
  • The most impressive out-performance against the S&P 500 index is from the Top 5 signal group in the 1-year horizon with 3.67 times higher return
  • Predictions reach up to 88% hit ratio regardless of macroeconomic conditions
  • Every signal group has hit ratios above 51% for all time horizons
  • I Know First provides an investment strategy for institutional investors that generated a return of 142.57% and exceeded the S&P 500 return by 118.93% for the analyzed period

Read Full Report

Bottom Up Stock Market Forecast for the main World Indices by using AI Powered Predictive Algorithm

Executive Summary

In this stock market forecast evaluation report, we introduced new bottom up signal generation methods using the I Know First AI Algorithm for the following world indices: S&P 500, DAX and CAC40. We compare the new bottom up signals’ performance with the direct forecasts’ performance generated for the indices using our short term model (1 to 6 days) and daily model (3 days to 3 months). Our analysis covers the time period from April 2, 2019, to April 3, 2020.

Bottom Up Forecast Evaluation Highlights:

  • Short term model showed higher hit ratio than the Daily model for the 3 days’ time horizon for all indices analyzed.
  • S&P 500 predictions showed higher hit ratio for all time horizons and methods than DAX and CAC40.
  • Every method has hit ratios higher than 55% for the index’s predictions using Daily Model.

Read more