FangQuant › Strategies

Quant Investment in China A-share market

Fang submitted 2017-06-26 15:31:27

1. China Equity Markets—Overview
Stocks Listed& Market Cap since 1990

2. Two Stock Exchanges—Shanghai and Shenzhen
1) Date Established:
Shanghai Stock Exchange:
1990.12
Shenzhen Stock Exchange:
1990.12
2) Stocks Listed(as of 2016.12.31):
Shanghai Stock Exchange:
28.46 Tr RMB
Shenzhen Stock Exchange:
22.30 Tr RMB
3) Type of Listings: Stocks,Mutual Funds,Bonds,Repos,Warrants
4) Main Function:
Provide trading location and facility for listed securities
Set up rules and regulations for the Exchange
Responsible for listings of IPOs
Supervision and surveillance for securities trading activies
Supervision and surveillance on exchange members and listed companies
Timely disclosure of information relating to listed securities

3. Fame-French in China by CSHW
1) Fama-French in China—Size and Value Factors in China Stock Returns,
2) Main Results—
● CSHW find a significant size effect but weak value effect in the China stock market
● In both time-series regression and Fama-Macbeth cross-sectional tests,SMBappears to be the strongest factor in explaining the cross-section of China stock return.
3) CSHW use the China Capital Market (CCM) Datebase provided by the China Academy of Financial Research(CAFR)—accounting data, historical A-share prices and returns.

4. Fama-French in China by CSHW
1) Formation of 10 size—and B/M-sorted portfolios
2) Size is measured as the floating A-share market capitalization at the end of June each t.
● Only floating A-shares are investable for general domestic investors, while non-floating shares or other types of floating shares such as B and H are not.
● Non-floating shares are not actively traded and their transaction prices are not determined in the open market but through private negotiations.
3) Book-to-Market ratio(B/M) is calculated as the fraction of book value of equity per share and floating A-share prices at the end of Dec in the previous year t-1.
4) The portfolios are kept unchanged for the following 12 months.

5. Factor-based Stock Selection: Our Approach
1) Universe:
● Benchmark: CSI300 or CSI500
● Select from Benchmark index constituents only
2) Rank:
● Decide factors to use
● Rank according to each factor
● Optimize among factors
● Generate composite score
3) Risk Control:
● Sector neutral
● Proper exposure on SIZE

6. Factors Pool:
Factors we monitor:
1) Valuation:
● P/B
● P/E: historical, Consensus Forecast
● P/Sales
● P/Cash
● EV/EBITDA
● Debt/Assets
● Etc.
2) Technical:
● Momentum
● Price Reversal
● Liquidity: Recent volume, Medium-term average volume
● Volatility: Return vol, Residual vol of Beta model
● Etc.
3) Other Style:
● Size
● Growth: Historical Operating income growth, Historical EPS growth, Consensus forecast of EPS growth
● Etc.

7. Factor Performance in A-share:300
1) Single Factor: Price/Book
● Definition: Price/Book Value Per Share; Rank from Low to High
● Measure: Top 10% ranked stocks& sectors neutralized: Compare against index
● Performance: Persistently outperform; High IR
2) Single Factor: Size
● Definition: Market Value; Rank from small to big
● Performance: Performance not satisfactory; suffered significant loss from time to time
3) Single Factor: Short-term Reversal:
● Definition: Total return of previous month; Rank from low to high
● Performance: Overall not good; Tough time in 2H 2014 as well as in 2016

8. Factor Performance in A-share: 500
What if we apply the same set of factors to smaller stocks?
1) 500 Single Factor: Price/Book
● Performance: negative in 2013; large draw-down in 1H 2015
2) 500 Single Factor: Size
● Performance: works well overall; much less affected by style change in late 2014 than in 300 universe
3) 500 Single Factor: Short-term Reversal
● Performance: similar as in 300 universe; suffered loss during the 2014-2015 market rally
4) 500 Optimized Portfolio
● Performance: Overall quite good, and much better than the 300 universe in 2016

9. Challenges:
1) Liquidity
Lack of liquidity is generally not good for alpha generation
● Low trading volume in 2012-2013 and 2016-2017
● Tough time for selecting stocks from benchmarks, but works relatively better using all-A sample due to size
2) Style Rotation
Style change: Elephant Dance
● Triggered by PBRC rates cut in Dec,2014, widely viewed as the start of a new monetary easing cycle
● Size, one of the most profitable factor in A-share, experienced double digit loss in one month
3) Trading Limits on Futures
● 2015: Bubble, and Burst
● New policy on stock index futures:
(1) Program trading suspended;
(2) Impose restricts on stock index futures: daily limits on speculative tradingaccounts;
(3) Trading commission on intraday trading ever raised to 100 times
● Consequence:
● Persistent negative basis between cash index and futures --> Market neutral strategy paralyzed
4) Disruption by National Team
● Since 2016, national team has been active in China A share market
(1) Circuit Break triggered twice in one week during Jan 2016;
(2) “National Team” to support the market-->distortion for quants
5) Love and Hate on Size:
With an all-A sample, size has proven to be a great source of excess returns
● Since 2012,size has delivered an average 40+% of annualized excess return over CSI500 benchmark
● However, size has experienced 3 major drawdowns( near 20% in the worst case)
● We have just experienced another large drawdown since Dec 2016

10. Opportunities: Machine Learning
With an all-A sample, Machine Learning has the potential to help mitigate the large drawdown from size exposure
● Adaboost is a machine learning algorithm which can rapidly adapt to changing environments
●It’s designed to dynamically select useful factors, hence suitable for a rapidly changing market conditions

11. Summary:
1) Quant strategies have investment value in China A-share market
● Factor models can generate much larger alpha than in other markets
● This is probably because of high participation by individual investors in A-share market
● However, market environments change regularly so models need to be adjusted accordingly
2) Hedging is challenging under current situation
● Negative basis is a sure loss while alpha is an expected return
● Restriction on futures trading and lack of other short tools caused the persistent negative basis
3) Market neutral strategies are in high demand by banks and HNW individuals in China



Copyright by FangQuant.com





Currently no Comments.