Misleading Effects of Backtest Overfitting in Investment Strategy

Backtesting is a process where a financial strategy is tested on a predictive model that is built based on historic data. The strategy is validated based on its performance on the model.

An article published in the Notices of the American Mathematical Society looks into this investment practice and highlights the perils of performing a large volume of backtest on a historical data set that it would skew the results and favor the data set but would not be as practical or efficient in another out of sample set.

This is known as backtest overfitting. Since the model is based on past financial portfolios and tweaked to single out the best performing stocks, it may not be an efficient one and may just end up highlighting the extremes and vagaries in the financial data.

The article, "Pseudo-Mathematics and Financial Charlatanism: The Effects of Backtest Overfitting on Out-of-Sample Performance" was written by David H. Bailey, Jonathan M. Borwein, Marcos Lopez de Prado, and Qiji Jim Zhu.

Founded in 1888 to further mathematical research and scholarship, today the more than 30,000 member American Mathematical Society fulfills its mission through programs and services that promote mathematical research and its uses, strengthen mathematical education, and foster awareness and appreciation of mathematics and its connections to other disciplines and to everyday life.

Report Studies Brain Process and Psychological Factors Behind Market Bubbles

A study led by researchers at the California Institute of Technology show that stock market bubbles are linked to traders' brain processes that predict how other traders will behave rather than relying on economic factors such as price and demand when making financial decisions.

Market bubbles are financial activities where stock prices are increasing above their value because of traders speculation that the price of the asset will steadily rise because of increased demand. Although there are two activities that are associated with a market bubble; the boom and the burst.

A market boom is the phase of the bubble where the asset price increases dramatically alongside with increased traders investment. A burst happens when the opposite occurs, prices sharply drop and there is a widespread drop in spending on the asset. Prices in an economic bubble can fluctuate erratically, and become impossible to predict from supply and demand alone.

The study offers a glimpse on the brain processes that factor in financial decisions and behavior leading to the formation of market bubbles.

The Stock Market - High Frequency Trading, the Algorithms and the Science Behind It

In Wall Street and other trading environments, some investors use sophisticated technological tools to trade securities like stocks or options. This is called High Frequency Trading (HFT).

HFT utilizes super computers and algorithms to generate automatic trades. One major factor for high frequency trading is that information and actual stock trends are picked up by these super computers in real time and based on the algorithm, react accordingly.

An algorithm is a an order of sequential procedures for performing calculations. It is a step-by-step series of procedures used for calculation, data processing, and automated decision making or reasoning.

Distinguishing Characteristics of High Frequency Trading

Since trading behavior is based on the information coming in, the algorithm programmed into the system, and proprietary (built in) trading strategies, HFT is highly quantitative. The computer just reacts to the data and is highly objective.

Due to the dynamic movement of the market, stock and investment positions are considered temporary and can change in a manner of seconds. HFT systems may trade into a stock and trade out of it almost immediately.

In terms of a net investment position, HFT systems have none. High-frequency trading firms do not employ significant leverage, do not accumulate positions, and typically liquidate their entire portfolios on a daily basis.