Quantitative Methods + Qualitative Analysis = Superior Returns

SummaryPhilosophy• Value Investing provides an investor the highest probability of success since historically it has provided the

best returns.• Almost any investment could be justified by an enthusiastic investor as value based on assumptions.

Therefore the approach needs a method in order to focus a value investor’s effort. Quantitative valuemodels achieves this. A similar approach as Asness’ AQR, Greenblatt’s Gotham Funds and Ibbotson’s ZebraCapital.

• Black boxes are prone to failure, so fundamental (traditional bottoms-up, qualitative) analysis should beapplied. This boosts returns and reduces risk.

Method• Proprietary quantitative models based on combining value with growth, dividends, momentum and on its

own are used. 6-17 factors are used in each model. Each model and each individual factor were backtestedwith positive results. 65 data points are used.

• The models were based on combining factors that academics and other practitioners have also tested anddisplayed in books and papers.

• Fundamental analysis is done on the model output. Only the best opportunities are selected.

Strategies• Flagship FatAlpha Active Strategy. Gross exposure: 100%. Net exposure 90%+ during 2012-15. From 2012-

2015, the strategy returned 157% vs 60% for market.• FatAlpha Market Neutral Strategy launched in September 2015 with a target gross exposure of 100% and

net exposure of 0%. This strategy uses a short quantitative model designed in 2015.

2

Searching for the best strategy

• Market Facts:1. Most Active Fund managers underperform2. Market Gurus aren’t any better3. Wall Street predictions are mostly wrong

What approach is the most successful based on both theory and practice?

• Fact: Many successful fund managers are/were value investors: • Warren Buffett (Berkshire Hathaway), Peter Lynch (Fidelity Magellan), Sir John Templeton

(Franklin Templeton Investments), Joel Greenblatt (Gotham Capital), John Neff (Vanguard's Windsor Fund), Benjamin Graham, Al Frank, etc.

• Fact: Academic work has concluded that value investing results in significant alpha:• Recent Nobel Prize Winner, Eugene Fama found that low price-to-book stocks outperfomed

the market and was disappointed to see that this performance could not be explained by the Capital Asset Pricing Model (CAPM).

• Professors Lakonishok, Shleifer and Vishny published "Contrarian Investment, Extrapolation, and Risk" that showed that low P/E, P/Bk and P/CF stocks outperformed the high P/E, P/Bk, P/CF stocks and the market averages.

• Dreman and Lufkin looked at 27 years of annual data to come to the same conclusion when studying Price to Earnings, Price to Book, Price to Cash Flow and High Dividend Yielding stocks. According to Dreman’s work not only do low PE stocks do better than high PE stocks but this outperformance occurs also during both positive and negative earnings surprises.

3

How to best implement a value investment strategy?

Quantitative Value Models have been backtested extensively with success and shows to have significant merit:

• O’Shaughnessy conducted an extensive in-depth study of market ratios and potential strategies in “What Works on Wall Street”. The conclusion was that buying stocks with low multiple or practically any combination of ratios led to market outperformance. He shows numerous multi-factors models that outperform the market. Survivorship bias and other potential statistical issues were accounted for.

• While at the University of Chicago, professor Piotroski wrote "Value Investing: The Use of Historical Financial Statement Information to Separate Winners from Losers" In the paper Piotroski showed that by combining a 9-factor binary model to filter the low Price to Book companies, one could outperform the market. FatAlpha backtested this idea and confirmed Piotroski's conclusion to be true.

• S&P Analyst Richard Tortoriello wrote "Quantitative Strategies for Achieving Alpha". Tortiellotested numerous single, two-factor, and three-factor models while providing a framework and logic for creating multi-factor models. Tortoriello backtests show that valuation, more than anything, drives superior returns and is the strongest of the quantitative factors. He recommends using value factors as the base of any model.

• Fund manager and Columbia professor Joel Greenblatt in his NYT bestseller backtests his ‘magic formula’ method over 17 years (1987-2004), and shows an average outperformance of about 10% per year vs the S&P 500.

4

Initial Backtests of Quantitative Methods

• FATALPHA conducted its ownbacktests of numerous valuequantitative models. The modelsfocused on company fundamentalsand market ratios.

• The results confirmed previous studies.The backtested results of two suchmodels are shown on the right andbelow.

• FATALPHA mainly uses such modelsto produce lists of potentialinvestments.

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Return

S&P 500* 20% 31% 27% 20% -10% -13% -23% 26% 9% 3% 14% 4% -38% 23% 13% 0% 5%

Value Model 11% 49% 1% 7% 8% 40% -4% 55% 46% 15% 26% -21% -38% 132% 41% 5% 18%

Multiple/Dividend Model 4% 33% 3% 1% 27% 21% -4% 26% 23% 14% 20% 11% -30% 52% 13% 6% 12%

0.0%2.0%4.0%6.0%8.0%

10.0%12.0%14.0%16.0%18.0%

20.0%

S&P 500 Multiple/DividendModel

Value Model

5%

12%

18%

1996 – 2011 Annual Backtested Return

* Based on Bloomberg ticker: SPX Index

5

Current Quantitative Models Used

1996 – 2013 Annual Backtested Return

* Based on Bloomberg ticker: SPX Index

6%

16%

13%

16% 16%14% 15%

17% 17%

0%2%4%6%8%

10%12%14%16%18%20%

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Return AlphaValue 1Q 21% 32% -1% 11% 8% 39% -9% 58% 25% 8% 20% -11% -38% 96% 28% -4% 17% 51% 16% 10%Value 1 12% 42% -6% -2% 0% 25% -18% 48% 28% 8% 25% -9% -42% 100% 21% 3% 13% 63% 13% 7%Value 2 16% 31% 4% 14% 7% 27% -10% 39% 30% 16% 19% -10% -38% 100% 23% 5% 19% 53% 16% 10%

T-Value 1 21% 27% 20% 8% 18% 15% 0% 36% 45% 18% 28% -6% -35% 47% 21% -2% 7% 49% 16% 9%T-Value 2 4% 32% 13% 7% 13% 18% -1% 34% 38% 23% 23% -11% -33% 43% 25% -2% 15% 49% 14% 8%Dividend 8% 28% 2% 11% 13% 23% -2% 44% 26% 18% 22% 6% -35% 59% 15% 5% 14% 43% 15% 8%Garp 1 26% 29% 2% 22% 13% 54% 3% 58% 21% 13% 12% 12% -37% 44% 27% -1% 13% 40% 17% 11%Garp 2 32% 31% -7% 28% 18% 47% 5% 51% 26% 13% 8% 6% -38% 63% 19% 0% 13% 44% 17% 11%

S&P 500* 20% 31% 27% 20% -10% -13% -23% 26% 9% 3% 14% 4% -38% 23% 13% 0% 13% 30% 6%

• The backtested models havethe following four themes:– Value,– Value with momentum,– Value with growth,– Value and dividends.

• Models contain between 6-17 factors

6

Robustness of Quantitative Value

• Value 1Q model ranks all the stocks in theRussell 3000.

• To test the robustness of this model, the3000 ranked stocks are broken into 10groups of 300 (i.e. 1-300, 300-600, 600-900, etc) and the returns of each havebeen calculated.

• As expected the top group outperforms,which each subsequent group registeringa lower return.

Decile 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Return1 21% 32% -1% 11% 8% 39% -9% 58% 25% 8% 20% -11% -38% 96% 28% -4% 17% 51% 16.0%2 21% 28% -2% 15% 17% 22% -11% 44% 22% 10% 15% -3% -42% 74% 28% -1% 17% 40% 13.7%3 21% 32% 1% 4% 17% 19% -7% 45% 20% 7% 17% -6% -35% 64% 24% -2% 15% 41% 13.2%4 18% 28% -2% 2% 11% 13% -10% 42% 19% 5% 17% -1% -38% 46% 26% -6% 19% 40% 10.8%5 14% 26% -1% 6% 13% 6% -13% 54% 17% 4% 18% -5% -33% 50% 21% -7% 20% 38% 10.7%6 20% 22% -3% 9% 13% 6% -20% 51% 17% 3% 17% -3% -34% 29% 20% -9% 14% 35% 8.6%7 18% 18% 4% 5% 8% -3% -22% 57% 16% 5% 16% 3% -36% 27% 22% -6% 14% 33% 8.0%8 15% 20% 3% 40% -8% -6% -31% 54% 13% 2% 9% 5% -38% 37% 24% -6% 18% 37% 7.9%9 10% 11% 11% 42% -9% -17% -40% 52% 6% 4% 11% 2% -41% 26% 25% -6% 13% 43% 4.8%

10 6% -4% 15% 70% -20% -30% -45% 51% 1% -1% 5% 9% -44% 22% 28% -4% 12% 35% 1.7%

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

14.0%

16.0%

1 2 3 4 5 6 7 8 9 10

16.0%

13.7% 13.2%

10.8% 10.7%

8.6% 8.0% 7.9%

4.8%

1.7%

1996 – 2013 Annual Return

7

Growth of $100,000 invested in models

• $100,000 become $300,093 in the S&P 500 vs $1,747,632 in model.• Portfolio rebalanced at year-end starting in 1995. Dividends not included.• S&P 500 benchmark used is SPX Index from Bloomberg

$1,436,330

$1,747,632

$1,360,729

$1,185,257

$300,093

$100,000

$300,000

$500,000

$700,000

$900,000

$1,100,000

$1,300,000

$1,500,000

$1,700,000

$1,900,000

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Value Value+Growth Value+Momentum Value+Dividends S&P 500

8

Value Proposition

EDGE 1: Value Investing Strategy • The most successful fund managers are value investors.• Academics conclude that value investing results in significant alpha.

EDGE 2: Quantitative Value Models• Ideas sourced using proprietary quantitative value models• Models backtested since 1996. • Approach tested by academics and market practictioners (eg. Fund manager Joel Greenblatt

and his ‘magic formula’ method described in his New York Times best seller)

EDGE 3: Qualitative Analysis • Overlaying bottoms-up fundamental analysis enhances returns and reduces risk.• Strategy beat stand-alone models in both 2013 (59% vs 51%) and 2014 (30% vs 4%)

ValueInvesting

FundamentalAnalysis

QuantitativeModels

9

Qualitative Analysis

• After producing short lists of potential investmentopportunities FATALPHA applies its qualitative overlay.

• During this process FATALPHA focuses on:– Strong Fundamentals

• Preference for strong cash flow generators• Low levels of debt (overleveraged companies are eliminated)

– Story• Cheapness could be due to specific reasons.• FATALPHA determines whether the market has exaggerated and if

there is a opportunity for either a short-term rebound or medium-termreversal.

– Technicals• Flattening to rising stocks are generally preferred, however stocks with

evidence of bottoming are also considered• Stocks in free-fall are generally avoided and placed on a watch list

10

Investment Methodology

Value Value + MomentumValue + Growth Value + Dividends

65 data points on each company in the Russell 3000.

~200 potential investment opportunities

Value Screening (every 2 weeks)

Filter out illiquid stocks (<$2.5m per day average volume)

Qualitative Fundamental Analysis

Apply sector concentration rules & compare to similar (if exist) in current holdings

Portfolio: 90-100% (0-10% discretionary), 5% per position (range 2.5-10%), ~18 names11

Why Value Investing & Models will continue work?

a) Value investing and models don’t always beat the market and as a result there isnot enough trust in them. Managers are not comfortable with this and quantsusually result in data mining historical data in search of the ‘holy grail’.

b) Investors won’t stick around to reap the long-term benefits. Fund managers arefocused on short-term results because they are judged on a quarterly and yearlybasis. Value models outperform over a period of 3-10 years. This timeframe istoo long for the majority of managers.

c) Value investing requires fund managers & investors to hold unpopular stocks andavoid the latest star stock. Many fund managers prefer to hold popular stocks formarketing purposes, peer-pressure and competitive reasons. Investments insectors and stocks which are out-of-favor are risky to their careers. It is easier toblame the market or bad luck for underperformance when holding blue chipsand market darlings. “No one ever got fired for buying IBM.”

d) Large funds are handicapped due to size to buy only very large companies andthus miss out on smaller investments. This is an investment opportunity for asmall portfolio such as FATALPHA.

12

The FatAlpha Advantage

1. The strategy has a strong framework and focus :• clear definition of what is value• definition does not shift based on manager preferences

2. Models on a stand-alone basis have historically outperformed.

3. Rigorous qualitative analysis results in understanding the ‘story’:• what has caused the drop in stock price• what factors will drive future returns

4. Fundamental and technical analysis avoids pitfalls of ‘black boxes’:• Common model problems avoided such as high sector concentrations• Positions are not blindly initiated – no risk of model failure due to data problems

(eg. errors, unadjusted, irregularities/abnormalities)• Value traps and “falling knives” averted

5. The portfolio is extremely liquid

6. Manager has “Skin In The Game” (invested personally in strategies).

13

Performance: FatAlpha Active

• Portfolio Beta: 0.91• Sharpe Ratio: 2.26 (vs 1.34 for S&P 500)• Benchmark used is the SPDR S&P 500 ETF (SPY) with

dividends after tax (15%) reinvested.• Monthly performance is shown above. Row labeled “Gross”

displays performance including broker fees only, while the row labelled “Net” displays performance including broker fees, and monthly management and performance fees of 1+10%.

• Table on left shows the months the S&P 500 fell and the corresponding performance of FatAlpha after fees.

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Year

2015Gross -0.83% 5.49% 1.84% -4.80% 3.11% -3.24% 1.14% -3.93% -2.39% 6.34% 2.64% -5.14% -0.63%Net -0.91% 4.95% 1.58% -4.88% 3.03% -3.32% 1.06% -4.01% -2.47% 6.25% 2.55% -5.22 -2.20%

S&P 500 (incl div) -2.51% 5.62% -2.01% 1.35% 1.29% -2.49% 2.64% -6.13% -3.01% 8.93% 0.39% -2.31% 0.81%

2014Gross -1.76% 7.64% 0.91% 5.08% 0.97% 0.83% 3.17% 3.67% -2.39% 6.21% 1.78% 0.86% 29.95%Net -1.84% 6.98% 0.74% 4.50% 0.80% 0.67% 2.78% 3.23% -2.47% 5.76% 1.53% 0.69% 25.51%

S&P 500 (incl div) -3.09% 4.55% 0.39% 1.06% 2.32% 1.58% -0.95% 3.95% -1.84% 2.74% 2.75% -0.80% 13.08%

2013Gross 6.45% 0.71% 5.35% 6.47% 4.19% -0.73% 7.45% -1.00% 2.78% 6.26% 5.50% 4.34% 59.01%Net 5.72% 0.56% 4.74% 5.74% 3.69% -0.82% 6.71% -1.08% 2.53% 5.55% 4.87% 3.82% 50.61%

S&P 500 (incl div) 5.72% 1.28% 3.34% 2.29% 2.36% -1.85% 5.60% -3.00% 2.66% 5.05% 2.96% 2.04% 29.60%

2012Gross 2.89% 9.12% 3.03% 0.66% 5.58% 1.79% 25.14%Net 2.52% 8.13% 2.65% 0.52% 4.95% 1.53% 21.88%

S&P 500 (incl div) 2.37% 2.51% 1.99% -1.36% 0.57% 0.18% 5.60%Since Inception Gross: 157% Net: 125% S&P 500: 60%

S&P DOWN MONTHS

Month Mo 1/10% S&P 500

Oct-12 0.52% -1.36%Jun-13 -0.82% -1.85%Aug-13 -1.08% -3.00%Jan-14 -1.84% -3.09%Jul-14 2.78% -0.95%Sep-14 -2.47% -1.84%Dec-14 0.69% -0.80%Jan-15 -0.91% -2.51%Mar-15 1.58% -2.01%Jun-15 -3.32% -2.49%Aug-15 -4.01% -6.13%Sep-15 -2.47% -3.01%Dec-15 -5.22% -2.31%

Average -1.28% -2.41%Beat 10 / 13 times

14

FatAlpha Active Vs Investment “Gurus”

Investment Gurus:Bill Ackman, Pershing SquareDavid Einhorn, Greenlight CapitalDan Loeb, Third PointWarren Buffett, Berkshire Hathaway

Footnotes:• Performance is after fees.• 2012 includes 3Q & 4Q only.

26.1%

8.8%

1.9%

12.8%13.9%

-20.0%

-10.0%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

2012 2013 2014 2015 per year

FatAlpha

Ackman

Einhorn

Loeb

Buffett

15

FatAlpha Active Vs Benchmarks

Benchmarks:Value Model is the backtested proprietary screen used to find the majority of investments. Performance of the topdecile is shown here. The results here prove that the qualitative analysis overlay adds value to the process.S&P 500 is the SPY with dividends after tax re-invested.IVE is the iShares S&P 500 Value ETF. Total return is shown.

26.1%

10.7%

14.4%13.7%

-20.0%

-10.0%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

2012 2013 2014 2015 per year

FatAlpha

Value Model

S&P 500

IVE

16

Portfolio Breakdown by Size of Market Cap

Large cap: great or equal to $10 billionMid caps: $2 – $10 billionSmall cap: $300 million - $2 billionMicro cap: less than $300 million

2015 2014 2013 2012Size Large Mid Small Micro Large Mid Small Micro Large Mid Small Micro Large Mid Small Micro

January 45% 27% 20% 0% 39% 18% 10% 0% 31% 27% 31% 12%February 62% 18% 14% 0% 66% 18% 7% 2% 35% 36% 21% 8%

March 46% 37% 12% 0% 53% 29% 14% 0% 42% 28% 30% 0%April 34% 43% 16% 0% 48% 30% 15% 0% 54% 17% 29% 0%May 38% 36% 21% 0% 38% 38% 20% 0% 68% 0% 32% 0%June 40% 30% 27% 0% 44% 35% 17% 0% 54% 0% 46% 0%July 40% 28% 25% 0% 67% 14% 14% 0% 58% 17% 24% 0% 16% 43% 0% 41%

August 34% 38% 26% 0% 58% 25% 12% 0% 64% 17% 18% 0% 18% 24% 38% 17%September 27% 40% 30% 0% 49% 33% 17% 0% 53% 27% 20% 0% 11% 33% 37% 19%

October 32% 39% 26% 0% 48% 37% 14% 0% 57% 29% 14% 0% 22% 30% 38% 10%November 51% 24% 21% 2% 51% 28% 11% 0% 75% 18% 7% 0% 21% 33% 32% 13%December 42% 28% 22% 3% 55% 18% 18% 0% 79% 16% 5% 0% 25% 36% 30% 9%

* Net remaining balance is in cash

17

Volume in USD of Portfolio Holdings

Used the three month average price and volume to calculated the liquidity of each stock.Four categories have been used: stocks which trade $20 million and above per day,between $10-20 million per day, between $2.5-10 million per day and stocks with dollarvolume below $2.5 million per day.

2015 2014 2013 2012Size >$20m $10-20m $2.5-10m $2.5m> >$20m $10-20m $2.5-10m $2.5m> >$20m $10-20m $2.5-10m $2.5m> >$20m $10-20m $2.5-10m $2.5m>

January 77% 5% 10% 0% 58% 7% 2% 0% 58% 13% 25% 4%February 82% 5% 6% 0% 81% 8% 2% 2% 65% 14% 18% 3%

March 82% 5% 6% 0% 80% 14% 2% 0% 69% 12% 19% 0%April 80% 7% 6% 0% 84% 7% 3% 0% 71% 14% 14% 0%May 78% 5% 11% 0% 80% 7% 3% 5% 84% 0% 16% 0%June 80% 5% 12% 0% 81% 8% 3% 5% 85% 0% 15% 0%July 78% 6% 9% 0% 83% 5% 2% 5% 92% 0% 8% 0% 59% 0% 41% 0%

August 84% 6% 8% 0% 83% 5% 2% 5% 83% 17% 0% 0% 49% 0% 48% 0%September 80% 10% 7% 0% 82% 4% 7% 5% 81% 7% 12% 0% 44% 5% 47% 4%

October 83% 9% 5% 0% 80% 5% 8% 5% 89% 0% 11% 0% 52% 9% 26% 13%November 88% 3% 6% 0% 74% 5% 6% 6% 96% 0% 4% 0% 54% 12% 25% 9%December 85% 3% 7% 0% 73% 9% 3% 6% 95% 0% 5% 0% 61% 12% 21% 6%

* The net remaining balance is in cash

18

Gross & Net Exposure

Definitions:

Long Exposure = market value of longs* / market value of portfolio (NAV)Short Exposure = market value of shorts* / market value of portfolio (NAV)

Gross Exposure = Long Exposure + Absolute value of Short Exposure

Net Exposure = Long Exposure + Short Exposure (as shorts are a negative %)

* Where options are present, the value used is the underlying exposure adjusted by delta (i.e. number of contracts * 100 * stock price * delta)

2015 2014 2013 2012Month Exposure Exposure Exposure Exposure

End Long Short Gross Net Long Short Gross Net Long Short Gross Net Long Short Gross NetJanuary 97% -5% 101% 92% 72% -5% 78% 67% 100% 0% 100% 100%

February 93% 0% 94% 93% 96% -3% 98% 93% 100% 0% 100% 100%March 94% 0% 94% 94% 96% -7% 103% 89% 100% 0% 100% 100%April 93% 0% 94% 93% 94% -5% 100% 89% 100% 0% 100% 100%May 95% 0% 95% 94% 95% -1% 95% 94% 100% 0% 100% 100%June 97% 0% 97% 97% 97% -1% 97% 96% 100% 0% 100% 100%July 93% 0% 93% 93% 95% -1% 95% 94% 100% 0% 100% 100% 100% 0% 100% 100%

August 98% 0% 98% 98% 95% -1% 95% 94% 99% -3% 102% 96% 97% 0% 97% 97%September 98% 0% 98% 98% 99% -1% 100% 99% 99% -1% 101% 98% 100% 0% 100% 100%

October 96% 0% 96% 96% 99% 0% 99% 98% 96% -6% 102% 91% 100% 0% 100% 100%November 98% 0% 98% 98% 95% -5% 100% 89% 96% -10% 106% 85% 100% 0% 100% 100%December 95% 0% 95% 95% 96% -5% 101% 91% 100% 0% 100% 100% 100% 0% 100% 100%

19

Disclaimer

These presentation materials are confidential and may not be reproducedor distributed by the recipient. Certain of the economic and marketinformation contained herein has been obtained from published sourcesand/or prepared by third parties. While such sources are believed to bereliable, FatAlpha does not assume any responsibility for the accuracy ofsuch information. This document is for informational purposes only. Theinformation contained herein is subject to change. However, we are underno obligation to amend or supplement this document. Investors areadvised to conduct their own independent research into individual stocksbefore making a purchase decision. Under no circumstances does anyinformation herein or in any FatAlpha material represent a solicitation,endorsement, or recommendation to buy or sell any security or otherfinancial instrument or any financial service, nor does it constituteinvestment advice. In no event shall FatAlpha.com be liable to any memberor guest for any damages of any kind arising out of the use of any contentmade available by any of the FatAlpha material. Past performance is a poorindicator of future performance.

20