Axioma AXCN4 China Equity Factor Risk Models

Axioma AXCN4 China Equity Factor Risk Models

Model Overview
Asset Coverage As of 2018, the model covers over 5,400 A-Share and B-Share securities
listed on the Shanghai and Shenzen Stock Exchanges. Stocks from Chinese
issuers listed in Hong Kong and other overseas markets are also covered.
Estimation Universe Includes assets with sufficient size and liquidity, using selection criteria sim-
ilar to those employed by major index providers. More granular, localized
rules are also applied on a per-market basis to filter certain exchanges, asset
types, etc. In early 2018, the estimation universe comprised 900 securities
on average.
Model Variants (4) Medium-horizon and short-horizon, fundamental and statistical
factor models are available.

Model History Daily history from January 1999 onwards.

Forecast Horizon Medium-horizon model: 3-6 months.

Short-horizon model: 1-2 months.
DRAFT

Estimation Frequency Factor exposures, covariances, and asset-specific risks estimated daily.

Axioma 1
Axioma AXCN4 China Equity Factor Risk Models

Fundamental Factor Model

Style Factors AXCN4-MH (11) AXCN4-SH (12)

Market-Based Factors
Market Sensitivity 2-year weekly beta versus the emerging mar- 1-year weekly beta
ket
Volatility 6-month average of absolute returns over Same (3-month average)
cross-sectional standard deviation, fully or-
thogonalized to Market Sensitivity
Short-Term Momentum N/A Cumulative return over past month
Liquidity Natural logarithm of the ratio of 3-month Same (1-month average daily vol-
average daily volume and 1-month average ume, 3-month Amihud ratio)
market capitalization, inverse of 6-month
Amihud illiquidity ratio, and proportion of
returns-traded over the last calendar year
Exchange Rate Sensitivity 2-year weekly beta to returns of a basket of Same (1-year weekly beta)
major currencies
Medium-Term Momentum Cumulative return over past year excluding the most recent month (same for MH
and SH models)
Size Natural logarithm of market capitalization (same for MH and SH models)

Fundamental Factors (same for MH and SH models)

Value Book-to-price, earnings-to-price and estimated earnings-to-price
Leverage Total debt (current and long-term liabilities) to total assets and total debt to
equity
Growth Realized sales growth, forecast sales growth, realized earnings growth, forecast
DRAFT

earnings growth
Profitability Return-on-equity, return-on-assets, cash flow to assets, cash flow to income,
gross margin, and sales-to-assets
Dividend Yield Ratio of sum of the dividends paid (excluding non-recurring, special dividends)
over the most recent year to average market capitalization

(See the AXCN4 Model Supplement Document for exact factor defini-
tions)

Industry Factors (41) GICS R -based industry classification with 0/1 assignments.

Local Factors (1) Meant to capture strong residual structure in certain markets not captured
by others factors. The model currently has only one such factor: Offshore
China.

Returns Model Uses a market factor as well as style and industry factors to model local
excess returns. The China market factor return, industry returns and style
factor returns are estimated from a single-stage root-cap weighted regres-
sion (with constraint on industries) of the daily excess local returns.

Returns History Medium-horizon model: 4 years of daily returns for factor correlations, 2
years of daily returns for factor volatilities.
Short-horizon model: 4 years of daily returns for factor correlations, 2 years
of daily returns for factor volatilities.

Estimation Robust linear regression using a Huber weight function and square-root
market capitalization weights.

Axioma 2
Axioma AXCN4 China Equity Factor Risk Models

Statistical Factor Model

Factor Structure 15 statistical factors.

Estimation 2-Pass Asymptotic Principal Components factor analysis with residual vari-
ance weighted returns.

Returns History One year of daily asset returns are used to estimate statistical factor expo-
sures. Four years of statistical factor returns are used to estimate statistical
factor covariances.

Factor Volatilities / Covariances

Estimation Covariance of exponentially-weighted daily factor returns.

Half-life Parameters Medium-horizon model: 125 days for variances, 250 days for correlations.
Short-horizon model: 60 days for variances, 125 days for correlations.

Autocorrelation Newey-West adjustment accounting for 3 days of autocorrelation. (a fixed

lag of 1 day is used for statistical factors)

Adjustments Axioma’s proprietary Dynamic Volatility Adjustment (DVA) procedure is

used to analyze trends in factor returns dispersion and adjust risk estima-
tion accordingly, to allow for heightened responsiveness in risk forecasts
and adaptability to the prevailing volatility regime.
DRAFT

Specific Risks
Estimation Variance of exponentially-weighted daily specific returns

History Medium-horizon model: 500 days.

Short-horizon model: 500 days.

Half-life Parameter Medium-horizon model: 125 days.

Short-horizon model: 60 days.

Autocorrelation Newey-West adjustment accounting for 1 day of autocorrelation.

Other Adjustments The Issuer Specific Covariance (ISC) captures covariances between secu-
rity lines of the same issuer using a cointegration model of price behavior.
Applies only to portfolios containing two or more securities from the same
issuer.

Axioma 3
Axioma AXCN4 China Equity Factor Risk Models

Data Deliverables
Availability Updated daily and downloadable via FTP and SFTP.

Historical Coverage Daily history from Jan. 1999 onwards.

Data Format Delimited text file (“flat files”) or proprietary database format for seamless
integration into Axioma Portfolio and Axioma Backtester R .

Benchmarks Local market benchmarks are available in a format compatible with Axioma
software products.

Exchange Traded Funds Broad coverage of regional, single country, and index linked ETFs. ETF
(ETFs) coverage for the model is determined by the model’s full coverage of the
underlying constituents in order to ensure consistency in the instrument’s
risk and exposure measures.

Pure Factor Portfolios Pure factor mimicking portfolios for Fundamental-MH and Fundamental-
(PFPs) SH models (daily update).

Model Descriptors Fundamental and market-based style factor descriptors for Fundamental-
MH and Fundamental-SH models (daily update).

Statistical Model Factor 250 days of Statistical factor returns history (daily update).
Returns (PRET)
DRAFT

Asset Identifiers Axioma ID, SEDOL,ISIN, local ticker, issuer/company ID.

Market Data Asset-level data including:

• Price, market capitalization

• 1-, 5-, 20-, and 60-day returns

• 5- and 20-day average daily volume
• Historical and predicted beta
Some items of market data may not be available in delimited text file for-
mat.

Axioma 4
Axioma AXCN4 China Equity Factor Risk Models

Appendix: AXCN4 Industry Factors vs. GICS R

GICS R Industry Groups (24) Axioma Industry Factors (41)

1010 Energy 101010 Energy ex Coal
101020 Coal & Consumable Fuels
1510 Materials 151010 Chemicals
151020 Construction Materials
151030 Steel
151040 Metals & Mining ex Steel
151050 Paper & Forest Products
2010 Capital Goods 201010 Aerospace & Defense
201030 Construction & Engineering
201040 Electrical Equipment
201060 Machinery
201070 Trading Companies Distributors & Conglomerates
2020 Commercial & Professional Services 202010 Commercial & Professional Services
2030 Transportation 203010 Transportation Non-Infrastructure
203050 Transportation Infrastructure
2510 Automobiles & Components 251010 Auto Components
251020 Automobiles
2520 Consumer Durables & Apparel 252010 Household Durables
252030 Textiles Apparel & Luxury Goods
2530 Consumer Services 253010 Consumer Services
2550 Retailing 255010 Retailing
3010 Food & Staples Retailing 301010 Food & Staples Retailing
3020 Food, Beverage & Tobacco 302010 Beverages & Tobacco
302020 Food Products
3030 Household & Personal Products 303010 Household & Personal Products
3510 Health Care Equipment & Services 351020 Health Care Equipment & Services
3520 Pharmaceuticals, Biotechnology & Life Sciences 352020 Pharmaceuticals, Biotechnology & Life Sciences
4010 Banks 4010 Banks
4020 Diversified Financials 402010 Financials ex Banks
4510 Software & Services 451020 IT Services
451030 Software
4520 Technology Hardware & Equipment 452010 Communications Equipment
DRAFT

452020 Technology Hardware, Storage & Peripherals

452030 Electronic Equipment, Instruments & Components
4530 Semiconductors & Semiconductor Equipment 453010 Semiconductors & Semiconductor Equipment
5010 Telecommunication Services 501010 Telecommunication Services
5020 Media & Entertainment 502010 Media
502020 Entertainment
5510 Utilities 551010 Utilities ex Renewable
551050 Independent Power & Renewable Electrivity Pro-
ducers
6010 Real Estate 601010 Real Estate

Axioma 5
Axioma AXCN4 China Equity Factor Risk Models

Appendix: AXCN4 Results Overview

Model Fit and Factor Performance

Figure 1: Average 1-month adjusted R-squared for the model estimation universe, 1999-2018. The results for
the medium- and short-horizon Fundamental models are very similar so only the results for the Fundamental-
DRAFT

MH model are shown.

Figure 2: Total risk bias statistics for selected benchmark portfolios, 2002-2018. The vertical lines represent
the bounds of the 95% confidence interval. Bias statistics beyond these lines are significantly different from
1.00.1

Axioma 6
Axioma AXCN4 China Equity Factor Risk Models

Figure 3: Total risk bias statistics for selected benchmark portfolios, 2002-2018. The vertical lines represent
the bounds of the 95% confidence interval. Bias statistics beyond these lines are significantly different from
1.00.
DRAFT

Figure 4: Cumulative return to each of the AXCN4-MH Fundamental Style factors, 1999-2018.

Axioma 7
Axioma AXCN4 China Equity Factor Risk Models

Figure 5: Cumulative return to each of the AXCN4-MH Market-based Style factors, 1999-2018.
DRAFT

Figure 6: Cumulative return to each of the AXCN4-SH Fundamental Style factors, 1999-2018.

Axioma 8
Axioma AXCN4 China Equity Factor Risk Models

Figure 7: Cumulative return to each of the AXCN4-SH Market-based Style factors, 1999-2018.
DRAFT

Figure 8: AXCN4-MH factors’ frequency of statistical significance, 1999-2018. Market Sensitivity, for exam-
ple, is statistically significant 42% of the time.

Axioma 9
Axioma AXCN4 China Equity Factor Risk Models

DRAFT

Figure 9: AXCN4-SH factors’ frequency of statistical significance, 1999-2018. Market Sensitivity SH, for
example, is statistically significant 43% of the time.

Axioma 10
Americas: 1-212-991-4500
EMEA: +44-203-621-8241
Asia Pacific: +852-8203-2790

New York Atlanta Chicago

Axioma, Inc. Axioma, Inc. Axioma, Inc.
17 State Street 400 Northridge Road 1 South Wacker Drive
Suite 2700 Suite 550 Suite 200
New York, NY 10004 Atlanta, GA 30350 Chicago, IL 60606
United States United States United States
Phone: +1-212-991-4500 Phone: +1-678-672-5400 Phone: +1-224-324-4279

San Francisco London Frankfurt

Axioma, Inc. Axioma, (UK) Ltd. Axioma Deutschland GmbH
201 Mission Street St. Clements House Mainzer Landstrasse 41
DRAFT

Suite #2150 27-28 Clements Lane D-60329 Frankfurt am Main

San Francisco, CA 94105 London, EC4N 7AE Germany
United States United Kingdom Phone: +49-(0)69-5660-8997
Phone: +1-415-614-4170 Phone: +44(0)-20-7856-2424

Paris Geneva Hong Kong

Axioma (FR) Axioma (CH) Axioma, (HK) Ltd.
19 Boulevard Malesherbes Rue du Rhone 69 2nd Floor Unit B, 17/F 30 Queen’s Road
75008, Paris 1207 Geneva, Switzerland Central Hong Kong
France Phone: +41-22-700-83-00 China
Phone: +33(0)1-55-27-38-38 Phone: +852-8203-2790

Tokyo Singapore Sydney

Axioma, Inc. Axioma, (Asia) Pte Ltd. Axioma (AU) Ltd.
Japan Tekko Building 4F 30 Raffles Place Level 17
1-8-2 Marunouchi, Chiyoda-ku #23-00 Chevron House 9 Castlereagh Street Sydney,
Tokyo 100-0005 Singapore 048622 NSW 2000 Australia
Japan Phone: +65-6233-6835 Phone: +61-(0)2-8074-3104
Phone: +81-3-6870-7766

Sales: sales@axioma.com
Client Support: support@axioma.com
Careers: careers@axioma.com