lecture 8: High-Frequency Trading, Temporal patterns & Cycles

Question 1

Market Making

Answer 1

an activity where a firm’s trader stands ready to buy and sell a particular stock on a regular and continuous basis at a publicly quoted price on a listed exchange

Question 2

What is High Frequency trading (HFT)

Answer 2

It is a type of trading using computer algorithms to rapidly trade securities

characterized by high speeds, high turnover rates, and high order-to-trade ratios using specialized order types, co-location, very short-term investment horizons, and high cancellation rates of orders

uses market making and proprietary trading strategies carried out by computers to move in and out of positions in milliseconds with high volumes and high speeds aiming to capture sometimes a fraction of a cent in profit on every trade using insignificant amounts of capital.

–> Given the short holding periods, HFT can potentiallyachieve Sharpe ratios tens of times higher than traditional buy-and-holdstrategies

Question 3

HFT firms perform “Market making” activities using what?

Answer 3

a set of high-frequency trading strategies that involve placing a limit order to sell (or offer) or a buy limit order (or bid) in order to earn the bid-ask spread

–> By doing so, they provide a counterpart to incoming market orders

Question 4

are HFT firms under obligation to maintain this activity during periods of extreme volatility?

Answer 4

no

Question 5

Impacts of HFT on liquidity

Answer 5

HFT has led to a reduction in bid-ask spreads and an increase in trading volume

However, trading volume and narrower bid-ask spreads may not be a reliable indicator of liquidity during times of significant market volatility

Question 6

Impacts of HFT on volatility

Answer 6

More aggressive HFT strategies may increase stock volatility

HFT Liquidity detection strategies “front run” ahead of large institutional orders amplifying price swings

Question 7

Front running

Answer 7

illegal practice of having knowledge of your client’s orders and executing your own orders first

Question 8

impacts of HFT on price discovery

Answer 8

Although HFT strategies are very rapid helping prices be more efficient to reflect new information in the short term, their effect on long term price discovery is less clear

HFT strategies are agnostic to a company’s fate and intrinsic value

Zhang (2010) shows that HFT hinders long term price discovery

Question 9

impacts of HFT on market confidence

Answer 9

Sophisticated “Algo” strategies and access to dark pools used by HFT firms give them an advantage over regular investors

Market events such as the flash crash of May 6, 2010 erodes confidence and create disincentives for individuals to invest in the market

Led to some market participants believing that the “markets are rigged,” with HFT having an edge at the expense of investors

Question 10

impacts of HFT on market confidence

Answer 10

Sophisticated “Algo” strategies and access to dark pools used by HFT firms give them an advantage over regular investors

Market events such as the flash crash of May 6, 2010 erodes confidence and create disincentives for individuals to invest in the market

Led to some market participants believing that the “markets are rigged,” with HFT having an edge at the expense of investors

Question 11

impacts of HFT on market confidence

Answer 11

Sophisticated “Algo” strategies and access to dark pools used by HFT firms give them an advantage over regular investors

Market events such as the flash crash of May 6, 2010 erodes confidence and create disincentives for individuals to invest in the market

Led to some market participants believing that the “markets are rigged,” with HFT having an edge at the expense of investors

Question 12

positives of HFT

Answer 12

HFT has made markets more liquid and decreased transaction costs

Question 13

negatives of HFT

Answer 13

Traders who post standing limit orders that cannot reflect changes in value due to news changes fast enough and lose to HFT (certain order types are bait to HFT)

Question 13

negatives of HFT

Answer 13

Traders who post standing limit orders that cannot reflect changes in value due to news changes fast enough and lose to HFT (certain order types are bait to HFT)

Question 13

negatives of HFT

Answer 13

Traders who post standing limit orders that cannot reflect changes in value due to news changes fast enough and lose to HFT (certain order types are bait to HFT)

Question 13

negatives of HFT

Answer 13

Traders who post standing limit orders that cannot reflect changes in value due to news changes fast enough and lose to HFT (certain order types are bait to HFT)

Question 13

negatives of HFT

Answer 13

Traders who post standing limit orders that cannot reflect changes in value due to news changes fast enough and lose to HFT (certain order types are bait to HFT)

Question 14

negatives of HFT

Answer 14

Traders who post standing limit orders that cannot reflect changes in value due to news changes fast enough and lose to HFT (certain order types are bait to HFT)

Question 15

straight up illegal impacts of HFT

Answer 15

HFT traders front run traders who are working large orders, making their trades more expensive.

–> Quote matchers profit from standing limit orders by trading ahead such orders by improving prices slightly

Question 16

Dark Pools

Answer 16

Private exchanges not accessible to the public, usually owned by a broker-dealer named as such due to their lack of transparency

Question 16

Dark Pools

Answer 16

Private exchanges not accessible to the public, usually owned by a broker-dealer named as such due to their lack of transparency

HFT are engaged in a technology arms race by employing faster computers, locating servers closer to exchanges (co-location), using algorithmic code, employing many types of orders and paying for high speed data feeds and dark pool access and being faster by milliseconds

Markets need to be slowed because fast HFT will make slower HFT go out of business, and this decreased competition will make the few firms in business increase spreads and thus transaction costs will rise making it more costly for all participants

Algorithmic trading by HFT poses systematic market risks; flash crashes, algos out of control, market terrorism in the wrong hands (Market volatility has increased since the proliferation of HFT)

Question 17

CYCLE ANALYSIS

Answer 17

attempts to find recurring major and minor peaks and troughs in price movement for better trade timing

By adding short, medium and long term cycles together the actual price activity can be forecasted.

Question 18

CYCLE ANALYSIS

Answer 18

attempts to find recurring major and minor peaks and troughs in price movement for better trade timing

By adding short, medium and long term cycles together the actual price activity can be forecasted.

Question 19

left or right translation

Answer 19

when a market is trending, the cycle peak tends to shift left or right depending on the direction of the larger trend

–> usually, In a trading range, cycles are fairly regular in that the market peaks half way through the cycle

Question 20

left or right translation

Answer 20

when a market is trending, the cycle peak tends to shift left or right depending on the direction of the larger trend

–> usually, In a trading range, cycles are fairly regular in that the market peaks half way through the cycle

Question 21

left or right translation

Answer 21

when a market is trending, the cycle peak tends to shift left or right depending on the direction of the larger trend

–> usually, In a trading range, cycles are fairly regular in that the market peaks half way through the cycle

This is consistent with the notion that in rising markets, prices should spend more time going up and in falling markets prices should spend more time going down

Question 22

2 ARGUMENTS AGAINST THE CYCLE CONCEPT

Answer 22

If prices behaved in pure cycles, like radio waves or tuning forks, the numbers would easily fit into mathematical formulas that would give us precise predictions similar to what we know about ocean tides and sunrises. This has not happened.

We have been unable to identify causes of specific cycles that many agree do exist in prices.

Question 23

The three qualities of a cycle

Answer 23

AMPLITUDE

PERIOD

PHASE

Question 24

AMPLITUDE

Answer 24

measures the height of a wave from peak to trough signifying the strength of a cycle

measured from TROUGH TO TROUGH as the tops tend to take more time to develop. Bottoms are more easily defined

Question 25

The PERIOD (length) of a cycle

Answer 25

the time spent between troughs

Question 26

AMPLITUDE

Answer 26

measures the height of a wave from peak to trough signifying the strength of a cycle

Question 27

how are cycles measured

Answer 27

measured from TROUGH TO TROUGH as the tops tend to take more time to develop. Bottoms are more easily defined

Question 28

how are cycles measured

Answer 28

measured from TROUGH TO TROUGH as the tops tend to take more time to develop. Bottoms are more easily defined

Question 29

There are four important principles to cycles:

Answer 29

SUMMATION

HARMONICITY

SYNCHRONICITY

PROPORTIONALITY

Question 30

SUMMATION

Answer 30

holds that all price movement is a simple addition of all active cycles

By combining each cycle & projecting forward, future price targets can be forecast

Question 31

HARMONICITY

Answer 31

means that there are waves within waves and that they are usually related

Adjacent cycles are often related by small whole numbers (usually 2 – sometimes 3), there is a constant ratio applied to cycles

Question 32

SYNCHRONICITY

Answer 32

refers to the tendency for waves of differing lengths to bottom at the same time

Question 33

PROPORTIONALITY

Answer 33

describes the relationship between cycle period and amplitude

Longer-term cycles (cycles having a longer period) should also have greater amplitude

There should be a proportional relationship between cycles of differing period

–> For example, if a cycle’s period is 40 days, it should have proportional amplitude that is 2x the amplitude of a cycle that has a 20 day period.

Question 34

VARIATION

Answer 34

states that the principles stated above are just strong tendencies and that they are not hard and fast rules. Some ‘VARIATION’ does occur

Question 35

NOMINALITY

Answer 35

states that there tends to be a nominal set of harmonically related cycles that affect all markets

Question 36

TRANSLATION

Answer 36

The reason that we study lows in cycle analysis is because longer and shorter cycles tend to synchronize at their lows.

–> Peaks, on the other hand, almost never synchronize.

Peaks should occur at the halfway period of the cycle.

–> For example, a 20 day cycle should have a peak 10 days from its last low. This rarely happens. (Coincidentally, this is the reason that we use the more stable trough to record our cycle lengths.)

—-> Peaks can occur earlier or later than the halfway point. Their location away from the center point is called TRANSLATION.

Question 37

A RIGHT TRANSLATION

Answer 37

when the peak is beyond the halfway point

If the trend is up (BULLISH) the cycle is said to translate to the RIGHT

If the peak had occurred at the halfway point, we would be suspicious about the continuation of the uptrend

Question 38

A LEFT TRANSLATION

Answer 38

when the cycle peak occurs before the halfway point

If the trend is down (BEARISH) the cycle is said to translate to the LEFT

Question 39

An INVERSION

Answer 39

occurs when a peak occurs where a cycle low is expected.

Question 40

Are economic phenomena that are not necessarily observable in commodity and stock prices. The theory that western capitalist economies have 50 – 60 year boom periods followed by a bust. (economic expansion followed by a depression)
The question asked in 2008/9-were we due for a depression?

KONDRATIEFF WAVES (K-WAVES)

Each wave is broken down into 4 seasonal periods (approx. 15 years each) with discernible characteristics:

Answer 40

Winter: a period of concern, fear, panic, depression

Spring: a period of fear of depressio4 n, fragile confidence

Summer: a period of growing confidence

Autumn: a period of increasing confidence, that turns into extreme confidence and euphoria

Question 41

34 YEAR HISTORICAL CYCLES

Answer 41

Historical data suggests that 34 year cycles, composed of 17 year period of dormancy followed by a 17 year period of intensity, also appear to exist

Question 42

FOUR YEAR or PRESIDENTIAL CYCLE (KITCHIN CYCLE)

Answer 42

The National Bureau of Economic Research showed that from 1796 to 1923, the US economy suffered a recession on average every 40 months or approximately every 4 years.

Some have argued that this follows the four year presidential cycle, but this phenomenon occurs in countries that do not have presidential elections every four years.

Today the four year cycle, from price bottom to price bottom, is the most widely accepted and most easily recognized cycle in the stock market. Occasionally it strays from four years, but only by a portion of the year.

Tops sometimes fail to occur as regularly as the bottoms do. They may occur before or after when their 4 year period would occur.

Question 43

ELECTION YEAR PATTERN

Answer 43

Stock market returns can be related as a function of the US presidential electoral timeline.

Equity markets are:
weak during the POST ELECTION & MID YEAR (Years 1 & 2)
strong during MID YEAR 3 and PRE ELECTION YEAR (4)

Question 44

SEASONAL PATTERNS

Answer 44

Are most apparent in agricultural prices and commodities such corn, hogs and oil.

–> This relationship is beginning to wane due to global trade

Allows the investor to profit from certain price trends that occur year after year. Though the price levels and the extent of their moves may vary – we can expect certain commodities to increase and decrease at certain times of the year.

–> For example:
Orange juice contracts have been profitable 74% of the time over the past 35 years for short trades to be initiated on June 4 and closed on July 1.

Question 45

‘Sell in May and go away’

Answer 45

an old stock market adage that refers to the tendency for the stock market to decline from May to September and rise from October to April

Question 46

JANUARY BAROMETER

Answer 46

an old stock market adage that states, ‘as the S&P goes in January, so goes the year’. Therefore, an up January will foreshadow a year of positive equity returns

Question 47

JANUARY EFFECT

Answer 47

another theory that states that small cap equities tend to outperform the broader market in the first few days of the new trading year due to investors buying back stocks that were sold for tax loss reasons at the end of the previous fiscal year.

–> This has not worked as well in recent years because arbitrageurs have essentially ‘arbed’ it out of the market