Quantitative trading uses data and models to make trading decisions. Algorithmic trading uses software to automate decisions or execution. They often work together, but a model, an automated order and an investment management service are three different things.
What quantitative trading is, and what algorithmic trading is
The distinction is useful when choosing a service or building a process. Quantitative describes the analysis behind a decision. Algorithmic describes how a decision or an order is implemented in software. Systematic trading follows a defined, repeatable process; automated trading carries out some or all of that process without a person entering every instruction. These terms overlap rather than forming mutually exclusive categories.
| Approach | Main question | Illustrative example |
|---|---|---|
| Quantitative trading | What does the evidence support, and how much risk should the position take? | A model estimates whether a price gap is unusual and sizes a possible position. |
| Algorithmic trading | How does software turn a decision into orders? | A program divides an order into smaller orders or responds to a model’s signal. |
| Systematic trading | Can the process be repeated and checked? | Entry, position size and exit criteria are specified before the trade. |
| Automated trading | Which step runs without a person stepping in? | At the close, the program sends an order that was approved earlier. |
A quantitative model can inform a decision that a person executes manually. An execution algorithm can also carry out a discretionary decision without selecting the investment itself. In its 2020 report, SEC staff describes a broad range of algorithmic tasks, from investment decisions to order timing and routing; there is no single universal use of the term.
Does algorithmic trading mean high-frequency trading?
No. A program that adjusts positions once a day can be algorithmic. High-frequency trading describes a particular kind of activity where speed and market infrastructure matter greatly. Neither a Python script nor a fast computer establishes that a strategy has an economic advantage.
From quantitative research to a working strategy
A useful research record lets someone reconstruct why the strategy was considered and what would invalidate it. Before automating execution, the research should address:
- The hypothesis: why the opportunity might exist, its horizon and the conditions in which it could stop working.
- The data: what was available at each historical date, how missing observations were treated and whether the universe includes failed or delisted assets.
- The evaluation: a separation between development and evaluation data, with a record of how many alternatives were tried.
- The costs: commissions, bid–ask spreads, slippage, financing and borrowing costs where relevant.
- The risk: position size, concentration, leverage, loss limits and the response when a limit is reached.
- The operation: monitoring, order records, reconciliation and a defined way to stop the system.
For example, if the model uses quarterly earnings, record the release time of each and rerun the test with the model allowed to read them only after that time, even if the quarter ended months earlier.
A historical simulation is an estimate under assumptions. It can look convincing because the model was repeatedly adjusted to the same data. A risk limit is an instruction to act, not a guarantee of the price at which an order will execute.
What to check in algorithmic trading software
Separate the strategy from the software that operates it. A useful evaluation asks to see the operating process on a paper account, including an error or an interruption: which orders are still open, which have filled and how the position is rebuilt.
| Check | Evidence to request |
|---|---|
| Broker and model compatibility | Which accounts, instruments and order types it supports, and how the model connects. |
| Risk controls | What happens before an order is sent and when a limit is breached. |
| Execution records | How rejections, partial fills, cancellations and duplicate instructions are handled. |
| Reconciliation | How the software compares its positions and results with the broker’s records. |
| Operations and costs | Who monitors it, what happens after a disconnect, and which data and broker fees are separate. |
Paper trading can help test a workflow, but it does not reproduce every live execution condition. Interactive Brokers explicitly notes that simulated order execution may differ from live trading: it simulates fills from the top of the order book, so a test cannot measure a large order’s effect on the market. Treat a paper result as operational evidence with limitations, not as proof of future investment results.
What a strategy needs to run live, and what each platform solves, is in the deployment guide; the checks before every order, and their order, in why algorithmic trading needs its own risk controls.
Quantitative and algorithmic trading from Chile
Being based in Chile and trading Chilean instruments are separate questions. A provider in Santiago may research or operate in international markets. Clarify which markets the service covers, who holds the account, who decides what to trade, and who runs the software. If the account is measured in dollars and spending needs are in pesos, the currency conversion also affects the result relevant to that person.
What does it take to get started? An account at a broker with an API, the market data each strategy uses, a computer or server that stays on while trading, and a program that turns signals into orders, checks each order against risk limits and reconciles positions with the broker. That program can be written from scratch, built on an open-source engine or taken from a ready-made platform: the deployment guide compares the options.
What costs should be considered? The broker’s commissions and market data, the software, the computer and the connection and, if the account is in dollars, currency conversion to and from pesos.
A software licence lets a user operate a tool; it does not by itself appoint an investment manager. A managed account requires its own mandate and responsibilities. A research subscription supplies analysis and does not execute orders. Athena, from Santiago, offers the three separately, and each answers those questions differently. In Carteras, each account is in the client’s name at Interactive Brokers and Athena manages the portfolio under an individual mandate. With Athena QS Console, the user installs the software, chooses the strategies and operates their own account. Research publishes monthly analysis reports and executes no orders.
For operating your own models
Athena QS Console is algorithmic trading software for Python models at Interactive Brokers, with risk controls, P&L and reconciliation. The user installs and operates the software. The product page explains its scope and plans.
Explore Athena QS Console →For investment research
Athena QS Studio supports the research process, from the economic cycle and sectors to S&P 500 companies. It serves a different task from trade execution.
Explore Athena QS Studio →For delegating management
Athena manages individual portfolios, with each account in the client’s name at Interactive Brokers.
Explore Carteras →Questions and answers
Are quant trading and algo trading the same?
They overlap. Quant trading describes the use of data and models in decisions; algo trading describes software implementing decisions or execution. Either can appear without the other.
Can algorithmic trading be done from Chile?
Yes. Being in Chile doesn’t limit the markets: what can be traded depends on the broker, whose name the account is in and which markets it covers. From Chile it is possible to open accounts at international brokers such as Interactive Brokers and trade US markets.
Sources
Newsletter
Stay up to date with Athena's publications and market insights.
Athena Quantitative Strategies is a quantitative investment firm based in Santiago, Chile: it manages portfolios, publishes its market analysis and offers the software it uses to trade and research. Athena Quantitative Strategies is not registered in the CMF's Register of Portfolio Managers and is not supervised by it. This content is educational and informational; it is not investment advice, a recommendation or an offer of securities. All investing involves risk, including loss of capital.
