We previously explored on Skinny on Options Data Science how some try to predict market moves. The segment from May 27, 2015, Probabilities of Market Moves tested conditional probabilities (eg. if the S&P 500 is up three trading days in a row is there a much higher likelihood that it will be up or down the next day). Dr. Data explained how all models rely on previously known data such as market prices. Last week in the segment on July 7, 2016, Predicting (Extrapolating) the Future Dr. Data explained the science based method of attempting to predict the future which is known as predictive modeling. Today’s segment is part 1 of a planned two part series. Mike uses his visual tools of Data Science to make the intimidating subject of regression algorithms and what information we can glean from them that is most important, easily understood.
A list of information used in models to help them predict future price was displayed in a table. The types of information listed were, historical price data, sector the underlying is in, near a yearly high/low, historical historical Implied Volatility (IV), past month’s price range, news out, near earnings, and outside of the 1 Standard Deviation price range. Dr. Data then noted that most researchers start with regression algorithms. They try to predict prices in the future, while minimizing any errors.
Using a series of graphs Mike contrasted two different models. One was much better at coming closer to the actual price. It’s error, the eventual actual price minus the predicted price was small. Dr. Data explained that the second model, in which the error level was much higher, was actually much better because it more accurately predicted direction. Mike demonstrated why direction is much more important. Predicting the magnitude of the move correctly is great but we can be way off on the magnitude and still make money as long as we get direction right. That’s the big problem with regression algorithms.
For more related information see:
Watch this segment of The Skinny On Options Data Science with Tom Sosnoff, Tony Battista and Dr. Data (Michael Rechenthin, Ph.D) as Mike explains the weakness of regression algorithms and the importance of direction over a lower magnitude of error and he does it in a way that is understandable to all tastylivers.
This video and its content are provided solely by tastylive, Inc. (“tastylive”) and are for informational and educational purposes only. tastylive was previously known as tastytrade, Inc. (“tastytrade”). This video and its content were created prior to the legal name change of tastylive. As a result, this video may reference tastytrade, its prior legal name.
tastylive content is created, produced, and provided solely by tastylive, Inc. (“tastylive”) and is for informational and educational purposes only. It is not, nor is it intended to be, trading or investment advice or a recommendation that any security, futures contract, digital asset, other product, transaction, or investment strategy is suitable for any person. Trading securities, futures products, and digital assets involve risk and may result in a loss greater than the original amount invested. tastylive, through its content, financial programming or otherwise, does not provide investment or financial advice or make investment recommendations. Investment information provided may not be appropriate for all investors and is provided without respect to individual investor financial sophistication, financial situation, investing time horizon or risk tolerance. tastylive is not in the business of transacting securities trades, nor does it direct client commodity accounts or give commodity trading advice tailored to any particular client’s situation or investment objectives. Supporting documentation for any claims (including claims made on behalf of options programs), comparisons, statistics, or other technical data, if applicable, will be supplied upon request. tastylive is not a licensed financial adviser, registered investment adviser, or a registered broker-dealer. Options, futures, and futures options are not suitable for all investors. Prior to trading securities, options, futures, or futures options, please read the applicable risk disclosures, including, but not limited to, the Characteristics and Risks of Standardized Options Disclosure and the Futures and Exchange-Traded Options Risk Disclosure found on tastytrade.com/disclosures.
tastytrade, Inc. ("tastytrade”) is a registered broker-dealer and member of FINRA, NFA, and SIPC. tastytrade was previously known as tastyworks, Inc. (“tastyworks”). tastytrade offers self-directed brokerage accounts to its customers. tastytrade does not give financial or trading advice, nor does it make investment recommendations. You alone are responsible for making your investment and trading decisions and for evaluating the merits and risks associated with the use of tastytrade’s systems, services or products. tastytrade is a wholly-owned subsidiary of tastylive, Inc.
tastytrade has entered into a Marketing Agreement with tastylive (“Marketing Agent”) whereby tastytrade pays compensation to Marketing Agent to recommend tastytrade’s brokerage services. The existence of this Marketing Agreement should not be deemed as an endorsement or recommendation of Marketing Agent by tastytrade. tastytrade and Marketing Agent are separate entities with their own products and services. tastylive is the parent company of tastytrade.
tastycrypto is provided solely by tasty Software Solutions, LLC. tasty Software Solutions, LLC is a separate but affiliate company of tastylive, Inc. Neither tastylive nor any of its affiliates are responsible for the products or services provided by tasty Software Solutions, LLC. Cryptocurrency trading is not suitable for all investors due to the number of risks involved. The value of any cryptocurrency, including digital assets pegged to fiat currency, commodities, or any other asset, may go to zero.
© copyright 2013 - 2024 tastylive, Inc. All Rights Reserved. Applicable portions of the Terms of Use on tastylive.com apply. Reproduction, adaptation, distribution, public display, exhibition for profit, or storage in any electronic storage media in whole or in part is prohibited under penalty of law, provided that you may download tastylive’s podcasts as necessary to view for personal use. tastylive was previously known as tastytrade, Inc. tastylive is a trademark/servicemark owned by tastylive, Inc.