Algorithmic Trading Types

Part 2

Today, we continue our post about Types of Algorithmic Trading. Our previous article talked about 3 of them: HFT, CTA, Short Term CTA. Today, we will cover Factor Trading and ETF Rebalance.


Factor models gained a lot of popularity in the early 2000s. Currently, about $2 trillion is managed by funds based on factorial models.

Factor models are arranged as follows. The 3,000 US stocks are broken down into factors that have proven to be good at predicting stock performance over the medium term. Next, quants try to predict the movement of stock prices using linear combinations of factors, as well as the so-called “genetic convolutions” obtained by calculations.

Stocks are ranked by values ​​of factors. The variance is studied, for example 10% of the bottom stocks and 10% of the top stocks, ranked by a factor or by their linear combination.

Then the undervalued shares are bought and the overvalued shares are sold.

Examples of factors include: size, sector, cost ratio, volatility etc.

Factor models can struggle during periods of money printing since, this money, when it enters the market, begins to act like a vacuum cleaner, and shares of companies are bought not on the basis of in-depth analysis, but simply because investors are overly liquid and looking for risk assets.


Automatic rebalancing of ETF portfolios has gained popularity over the past 10 years. At the same time, the volume of issued ETFs in the world has increased to $7.5 trillion, and the share of long-term investments in the form of ETFs in the United States reached 40%, in Europe 20%.

The Robo-advisors automatically calculate new ETF portfolio allocations and give recommendations for rebalancing to the client, or perform it automatically.

Typically, such rebalancing has three main objectives:

  • Tax optimization (sale of unprofitable assets at the end of the tax period);
  • Targeting a given volatility in accordance with the client’s risk profile;
  • Maintaining the weights of different asset classes.

The algorithms used are quite complex and require serious qualifications; trading systems are also quite technological. The average annual return of the top 24 Robo-advisors over the past two years ranged from 5% to 7% per annum (risky portfolios ) and 2% -4% per annum for portfolios with fixed income.

On average, Robo-advisors are more effective than human financial advisors.

The volume of funds managed by this technology is $1 trillion. It is expected to grow to $1.5 trillion by 2023 and the number of clients will increase to 150 million.

Access for free ETF Quant portfolios

arty ETF portfolios are based on AI