HFT & Advanced Concepts
Chapter: 6.5 HFT & Advanced Concepts Level: 1 — Recognition Time Limit: 20 minutes Total Marks: 30
Section A — Multiple Choice (1 mark each)
Choose the single best answer.
Q1. High-frequency trading (HFT) is primarily characterized by:
- A) Holding positions for several months
- B) Extremely high order-to-trade ratios and very short holding periods
- C) Manual entry of large block trades
- D) Long-term fundamental analysis
Q2. "Colocation" in the context of HFT refers to:
- A) Placing traders in the same office building
- B) Positioning trading servers physically near the exchange's matching engine
- C) Combining multiple orders into one
- D) Sharing profit between two firms
Q3. Latency arbitrage exploits:
- A) Differences in dividend payments
- B) Tiny time delays in price information reaching different venues
- C) Currency exchange rate spreads
- D) Interest rate differences between countries
Q4. A TWAP execution algorithm slices an order based on:
- A) Trading volume distribution
- B) Equal intervals of time
- C) A fixed percentage of market volume
- D) The bid-ask spread size
Q5. A VWAP algorithm aims to execute at a price close to the:
- A) Time-weighted average price
- B) Volume-weighted average price
- C) Opening auction price
- D) Closing bid price
Q6. A POV (Percentage of Volume) algorithm:
- A) Trades a fixed number of shares every second
- B) Participates at a set percentage of ongoing market volume
- C) Trades only at the volume-weighted average price
- D) Executes the entire order at the open
Q7. A circuit breaker on an exchange is designed to:
- A) Cut electrical power to servers
- B) Temporarily halt trading when prices move sharply
- C) Speed up order matching
- D) Reduce broker commissions
Q8. Direct Market Access (DMA) allows a client to:
- A) Trade only through a broker's discretion
- B) Send orders directly to the exchange order book using broker infrastructure
- C) Access insider information
- D) Bypass all exchange rules
Q9. Smart Order Routing (SOR) is used to:
- A) Guarantee the highest possible price
- B) Route orders across venues to obtain the best execution
- C) Predict future earnings
- D) Automatically pay dividends
Q10. The "Flash Crash" of May 6, 2010 is most associated with:
- A) A sudden, extreme drop and rapid recovery in US markets
- B) A year-long bear market
- C) A currency devaluation
- D) A bank interest rate cut
Section B — Matching (1 mark each, 6 marks)
Q11–Q16. Match each term (left) with its correct description (right).
| # | Term | Description | |
|---|---|---|---|
| Q11 | Market making at scale | A | Participate at X% of traded volume |
| Q12 | Latency arbitrage | B | Continuously quoting bid/offer across many symbols |
| Q13 | Colocation | C | Halt trading during extreme moves |
| Q14 | POV algorithm | D | Profiting from speed differences in data |
| Q15 | Circuit breaker | E | Splitting orders across venues for best fill |
| Q16 | Smart order routing | F | Servers hosted next to the matching engine |
Section C — True / False with Justification (2 marks each, 14 marks)
State True or False (1 mark) and give a one-line justification (1 mark).
Q17. HFT firms generally profit more from large per-trade margins than from high trade volume.
Q18. Regulators such as the SEC and SEBI have expressed concerns that HFT may increase systemic risk and reduce market fairness.
Q19. A VWAP algorithm typically front-loads all its trading in the first minute of the session.
Q20. Colocation reduces latency because signal travel time is affected by physical distance.
Answer keyMark scheme & solutions
Section A (10 marks)
Q1 — B. HFT is defined by very short holding periods and huge order-to-trade ratios; other options describe slow/manual/long-term styles. (1)
Q2 — B. Colocation = placing servers physically inside/next to the exchange datacentre to cut latency. (1)
Q3 — B. Latency arbitrage profits from being faster to react to price updates arriving at different venues. (1)
Q4 — B. TWAP = Time-Weighted Average Price; slices are equal over time intervals. (1)
Q5 — B. VWAP targets the volume-weighted average price benchmark. (1)
Q6 — B. POV participates at a set percentage of the market's traded volume. (1)
Q7 — B. Circuit breakers halt trading temporarily on sharp price moves to restore order. (1)
Q8 — B. DMA lets clients place orders directly on the book via the broker's membership/infrastructure. (1)
Q9 — B. SOR routes orders to venues offering best execution (price, liquidity, cost). (1)
Q10 — A. The 2010 Flash Crash was a sudden extreme intraday plunge and rapid recovery. (1)
Section B (6 marks)
| Q | Answer | Why |
|---|---|---|
| Q11 | B | Market making = continuous two-sided quoting across symbols |
| Q12 | D | Latency arbitrage = profiting from speed differences |
| Q13 | F | Colocation = servers next to matching engine |
| Q14 | A | POV = participate at X% of volume |
| Q15 | C | Circuit breaker = halt trading in extreme moves |
| Q16 | E | SOR = split/route across venues for best fill |
(1 mark each = 6)
Section C (14 marks)
Q17 — False. HFT margins per trade are tiny; profit comes from very high volume of trades. (True/False 1 + justification 1)
Q18 — True. Regulators cite concerns over systemic risk (e.g., flash crashes) and fairness/speed advantages. (1 + 1)
Q19 — False. VWAP distributes trading according to expected volume profile across the day, not front-loaded. (1 + 1)
Q20 — True. Signals travel at finite speed; shorter physical distance → lower propagation latency. (1 + 1)
[
{"claim":"Section A has 10 one-mark MCQs totaling 10 marks","code":"section_a = 10*1\nresult = (section_a == 10)"},
{"claim":"Section B has 6 one-mark matching items totaling 6 marks","code":"section_b = 6*1\nresult = (section_b == 6)"},
{"claim":"Section C has 7 two-mark items totaling 14 marks","code":"section_c = 7*2\nresult = (section_c == 14)"},
{"claim":"Total paper marks equal 30","code":"total = 10*1 + 6*1 + 7*2\nresult = (total == 30)"},
{"claim":"Total question count is 20","code":"count = 10 + 6 + 4\nresult = (count == 20)"}
]