Take Your Investments to the Next Level with Kai

Kai Score Performance

as of May 1, 2017 | Kavout Corporation

Kai is Kavout’s artificial intelligence and machine learning model which provides predictions on stock performance for a period of up to six months. Kavout’s proprietary Kai Score, calculated by Kai, analyzes over 200 indicators and features to look for correlations that are important for predicting performance. It also analyzes each stock in terms of its quality, valuation, growth and momentum. A high Kai Score (eg. 9) assigned to a stock indicates a greater probability of outperformance whereas a low Kai Score (eg. 1) indicates a lower probability of outperformance.

Kai’s Performance vs S&P 500

Statistics1 Alpha Volatility Beta CAGR Sharpe
Top Picks 9.07% 12.42% 0.86 22.53% 1.71
S&P 500 0.00% 11.56% 1.00 14.23% 1.21
Bottom Picks -11.17% 19.56% 1.20 3.90% 0.29

Methodology

The Kai model uncovers the intrinsic relationships among the changes in the time series of over 200 features, making informed judgments on classifications and regressions. The model applies the most advanced AI technology: deep learning. Kai analyzes and interprets millions of parameters each day and is powered by GPU, which enables fast and powerful parallel computing.

If a company’s recent quarterly statements and price data were input into the Kai model, the model would make a judgment on the company’s probability of outperforming the market. A stock with higher possibility of outperforming the market would get a higher Kai Score, while a stock with higher possibility of underperforming the market would get a low Kai Score.

The sample data used in the test was price data and financial data of stocks of S&P 500 from January 2012 to May 2017. The stocks were put into different portfolios based on their Kai Score. The portfolio was rebalanced each month. The stocks were bought on the first trading day of each month and sold at the last trading day of same month. The transaction cost was set to 0, because the stocks in this sample are of high liquidity.

Features Used

Kai uses more than 200 features to build a deep learning neural net to calculate the Kai Scores. Choosing informative, discriminating and independent features is critical to constructing an effective neural network, which consists of millions of features. The Kai model uncovers the intrinsic relationships within the changes in the time series of over 200 features, making informed judgments on classifications and regressions. Kai analyzes many years of historical data when assessing these features. Examples of these features include:

•Financial statement information including sales, net income, and EBITDA.
•Scores including the Z-Score, F-Score and M-Score.
•Ratios including capital to long-term debt, ROA and ROE.
•The time series of the features above as well as price-related data.

Quality, Value, Growth and Momentum Ranks

Quality

This rank is used to determine the quality of stock. A high quality stock is indicative of a well- managed company with strong financial strength. This can include:

• Working capital to long-term debt

• Greenblatt ROC

• Dividend payout

Value

A high value rank indicates that a stock is underpriced. This can include:

• Earnings yield

• Price to book

• Enterprise value to EBITDA

Growth

Indicative of a stock’s upward trend and growth. This can include:

• ROA/ROE three year growth rate

• Sustainable earnings growth rate

Momentum

A high momentum is indicative of a better historical return. This can include:

• Relative strength index

• 52-week high/low

1Alpha is a measure of performance on a risk adjusted basis, the benchmark used is the S&P 500 Index.

Annual Volatility is a statistical measure of the dispersion of return for a given allocation and is measured as the standard deviation of the allocation's arithmetic return.

Beta is a measure of the volatility in comparison to the market as a whole. The index used is the S&P 500 Index.

The compound annual growth rate (CAGR) is the mean annual growth rate of an investment over a specified period of time longer than one year.

A maximum drawdown (MDD) is the maximum loss from a peak to a trough of a portfolio, before a new peak is attained. Maximum Drawdown is an indicator of downside risk over a specified time period.

The Sharpe ratio is a return/risk measure, where the return is defined as the incremental annual return of an investment over the risk free rate. The risk is defined as the standard deviation of the portfolio.

Disclaimers

This material is for informational purposes only and does not constitute an offer to sell, a solicitation to buy, a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory or other services by Kavout Corporation.

This information is not investment or tax advice. Kavout Corporation is not an investment adviser or a broker dealer. Investing in securities involves risks, and there is always the potential of losing all your money. Before investing, consider your investment objectives and speak with a professional.

Past performance does not guarantee future results, and the likelihood of investment outcomes is hypothetical. This is not an offer, solicitation of an offer, or advice to buy or sell securities.