How Do These Models Work
- Analyzes the top 100 ETFs by volume according to etfdb.com
- Uses the machine learning algorithms of the WEKA package from University of Waikato, New Zealand
in particular the RandomForest classifier.
- Uses the following stock technical analysis tools from ta-lib libraries by TicTacTec LLC.
to create various attributes.
- 3 Simple Moving Average: Periods 5, 13 and 19, days/weeks.
- Bollinger Bands
- Uses the following classifications:
- strong buy - when the price change is greater than the average positive change
- buy - price change > 0
- strong sell - when the price change is less than the average negative change
- sell - price change < 0
- What is shown in report:
Probabilities are computed using the confusion matrix.
'Results for Previous Reports' Notes
- Starting price data is selected using data from the next market day.
- All price data for a day is an average of the open, high, low and close for that day.
- Percentage change is computed using the final date's price divided by the starting price.
- When a correct call is made
If a buy call was made and is correct then the symbol's related percentage change is used in the average.
- If a sell call was made and is correct then the inverse of the percentage change is used in the average.
- When the call is incorrect
- If a buy call was made and is incorrect then the symbol's inverse percentage change is used in the average.
- If a sell call was made and is incorrect then the percentage change is used in the average.
- Market Percentage Change is defined as the
average of both the DJIA and SP500 for the reporting period.
©2017 - 2020 McVerry Report LLC Raleigh NC USA