Algorithm decision making

Question 1

Algorithm trading:

Answer 1

• automated trading
• involves buying and selling securities electronically to capitalise on discrepancies on
  stocks in other different markets
• a formula( algorithm) is used to determine whether to buy or sell the assets
• looks at prices at all the different stock markets
• parameters underlying the algorithm will need to be derived using data from
  appropriate sources

Question 2

Benefits of algorithm trading

Answer 2

• leads to quicker decisions being made
• leads to more consistent decisions being made
• leads to fairer decisions being made
• increased speed and efficiency of trading
• can result in low dealings costs on trades
• can facilitate the execution of complex trading strategies which were not possible
  previously

Question 3

Risks relating to algorithm trading

Answer 3

• risk of algorithm not performing as expected
• risk of making consciously or unconsciously decisions for some individuals that were
  unfairly biased ( decisions that materially impact people’s lives )
• risk of error in algorithm
• risk of data used to parametrise the model could be wrong – potential losses
• risk of algorithm not operating properly in adverse conditions
  o in very turbulent conditions, trading in markets may be suspended before the
  algorithm trade can be completed
• main risk is the possible impact on the financial system

Question 4

Areas of work in which data is used

Answer 4

➢ administration
➢ marketing
➢ premium rating - product pricing, determining contributions
➢ setting provisions
➢ experience analysis
➢ investment decisions
➢ accounting
➢ risk management – including using underwriting and reinsurance
➢ management information

Question 5

Benefits of data being controlled in one single integrated system

Answer 5

• reduced chance of existing data being corrupted
• reduced chance of inconsistent treatment of information
• likely better control over those that may enter or amend information