Institutional Trading
• What problems did the institutional buy and sell side have
Buy Side Problems:
1) Institutional supply and demand isn't represented on most exchanges or markets
2) When large orders are executed on exchanges, the institutional demand is paired
against retail supply, which results in increasing market impact (Traders have difficulty
moving large blocks of shares quickly and efficiently)
3) When they decide to trade, they need a way to quickly implement their decisions
before information leakage impacts their potential return
4) Also, with large size orders, they need to deal with decimalization, small trade sizes
(need more executions), fragmented markets (multiple competing venues), less likelihood
of broker capital management, need to quantify trading costs and best execution
Sell Side Issues:
1) How to keep their trading secret
2) How to minimize market impact
3) What is the sense of urgency
4) What if the market is moving and I miss it
5) How is my performance meaure
• What are the costs in executing a trade
- Commissions (institutional) and fees (less than 1¢ per share)
- Spread (The round trip cost from the bid/asked spread)
- Market impact – the order is information
- Opportunity cost (If I don’t complete the trade/If I miss a market move – delay)
- Infrastructural overhead (allocated)
• What is market impact
Market Impact is the effect that a market participant has when it buys or sells an asset.
Price fluctuates if a large buy order is transacted.
• What are a trader’s concerns
1) How to keep their trading secret
2) How to minimize market impact
3) What is the sense of urgency
4) What if the market is moving and I miss it
5) How is my performance meaure
• What is TCA
Transaction Cost Analysis, a pre-trade analytics tool, is used by more than 90% of the
buy side and hedge funds firms. Its purpose is to maximize performance by best
execution and to measure and have continuous marginal improvement.
• What are TCA’s inputs and outputs
Inputs: Asset’s trading history (historical trading data & real-time data), Current Market
Conditions, and Portfolio Manager’s urgency
Outputs: Cost prediction, projected market impact and strategies for cost reduction
• What are examples of trader (execution quality) benchmarks
- Decision price
- Previous day close
- Market open, high, low, close or average of these prices
- Arrival price (at trader’s desk)
- Daily or interval VWAP
• What is the VWAP, decision price , arrival price benchmarks
VWAP is the volume weighted average price of a security which is calculated on by
taking the sum of number of shares at a particular price for all trades and divide that by
total number of shares traded that period
When deciding to buy or sell stocks during portfolio construction, a portfolio manager
looks at the prevailing prices which are called decision prices
Arrival Price Benchmark - measures the stock price at the time of the order, throughout
the day, and at the time of execution.
• What are the system prerequisites for algo trading
1) Being connected
- At the buy side OMS, for the buy side trader
- At the sell side OMS, for the sell side broker
- At the both sides -- market place connectivity (via FIX instructions from the trader)
2) OMS with STP and FIX connectivity
3) Direct Market Access
4) Low latency market data – quotes
5) High capacity communications and trading infrastructure
• What is algo trading
Algorithmic trading involves execution strategies that seek liquidity, lower market impact
by breaking up orders into small sizes and minimize cost of trade
• Identify 2 types of algo trading
TWAP : Trade at even rate
VWAP
Matches the volume-weighted average price (VWAP) for the day or a user time period
Has parameters such as % volume to decrease the risk of the Institutional Trading rule
itself affecting the price of the security
• What is DMA
Direct Market Access is high access to exchanges, ECNs, ATS’s by smart order routing
and aggregation of fragmented markets
• What is naked access
Some high speed traders don’t want the delay associated with a stop at their broker’s
system and request naked access. Sponsoring firm lets them go directly to the market and
give broker’s market id. They bypass the broker’s risk system but broker is still
responsible for trade.
• Explain and evaluate a VWAP trade, what parameters might you set
VWAP trade matches the volume-weighted average price (VWAP) for the day or a user
time period. It is considered a better algo strategy than the TWAP because stocks trade in
a regular pattern with higher volume at open and close and the schedule is based on each
stock’s historical tick data. Also, it sends orders to the market in proportion to the trading
volume and dynamically tracks volumes with order sizes.
I could set the following parameters:
- % volume to decrease risk of the rule itself affecting price of the security
- Avoid predictable schedules and behavior that could be detected and exploited –
randomize size and time
• How is volume participation used
Volume participation refers to trades in line with a stock’s volume to the % of volume
you wish to be. Unlike the VWAP rule, which will slow down or accelerate the execution
of an order based on a pre-defined schedule, Volume Participation dynamically adapts 54
to volume conditions throughout the day.
• What are the unintended consequences of algo trading
- further reduction in liquidity
- Disintermediation of the sell side trader/broker
- Bypassing soft dollar arrangements
- Bifurcated market
> Small trades on exchanges, ECN’s, etc.
> Blocks on dark ATSs, etc.
- Impact of historical data quality on pre-trade analytics and algo construction
• Explain stat arb
Statitical Arbitrage refers to attempting to profit from pricing inefficiencies identified
with mathematical models. Statistical arbitrage attempts to profit from the probability that
prices will move toward an historical average. Unlike ideal arbitrage, statistical arbitrage
has risk.
• What potential issues exist in algo trading
- Quality of real time and historical data
- Capacity and speed for baskets
- Ability to aggregate market liquidity and route quickly
- Need to maintain anonymity
> Avoid mechanically timed orders
- Can a system learn and improve?
> Minimize model risk
- STP to manage orders thru settlement
- Independent or broker supplied?
- Who is responsible when The Box fails?
