liveBook · Manning

Portfolio management

As explained in Applied Reinforcement Learning

The ongoing process of choosing and adjusting investments to balance expected return, risk, and costs over time.

Portfolio management turns market views into holdings, balancing expected return against risk, diversification, and trading costs. In a systematic strategy, it is the portfolio-construction step between generating signals and executing orders. The decision is sequential: today’s allocation affects tomorrow’s outcome. section 12.3

Table 12.1 The portfolio-management decision as a Markov decision process
Element
In This System
State
Market price features, a news-signal block, portfolio holdings, and execution/risk context (63 numbers in total)
Action
One of eight named allocation templates: cash-heavy, defensive, equal-weight, growth/tech, energy, financials, broad risk-taking, or risk-avoiding
Reward
Daily portfolio return minus transaction costs and penalties for turnover, drawdown, and volatility
Episode
A window of 30 consecutive trading days
Transition
Advance one day, realize the next day’s return, and update portfolio holdings, value, and drawdown

The decision can be framed as a daily reinforcement-learning task: observe market and portfolio context, choose an allocation, then account for returns and costs. section 12.3

A daily allocation agent

A small agent uses market features and signals extracted from financial news to choose among eight allocation templates. It is designed for periodic portfolio decisions, not microsecond trading. section 12.3

How Applied Reinforcement Learning covers it

How deep it goes
Maps daily decisions to states, allocation actions, rewards, episodes, and transitions, without detailing execution mechanics.
Examples
Uses a daily agent choosing among cash-heavy, defensive, equal-weight, sector-focused, broad-risk, and risk-avoiding allocations.
What it assumes
Assumes trading is familiar and requires no separate domain lecture.
What you can do after
Build a daily allocation agent that uses market and news signals to choose holdings and accounts for returns, costs, and risk penalties.
How it connects
Connects portfolio construction with signal generation, systematic trading, execution, and reinforcement learning.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Applied Reinforcement Learning. The section links point to the parts of the book it draws on.