There has been a marked uptick in the use of TCA (Transaction Cost Analysis) services within the FX market over the last few years. Doubtless significantly inspired by the FX Global Code and MiFID II, the use of TCA in FX has been suggested in some quarters to be a box-ticking exercise (although this charge is similarly levelled at TCA in other asset classes too). This is unfortunate, and maybe it’s time for a name change. The very name ‘TCA’ suggests an after-the-fact analysis which may never see the light of day again. Implicit here is the need for ‘pre-trade TCA’: helping traders intelligently decide the most appropriate execution algo for the required objective (how to execute) and on which venue (where to execute). Note that we use ‘execution algo’ here in the broadest sense, covering the method of order execution in general; whether using market, limit and other order types or true execution algos such as TWAP. We use ‘venue’ to include ECNs, banks and non-bank liquidity. Whatever the requirements are for regulatory, or indeed client reporting, surely the most important aspect of TCA (or ‘execution analysis’) is the use of post-trade analytics to inform and improve current and future (pre-trade) order execution.

Optimising execution quality means performing ongoing execution analysis in real-time to enable traders to make informed decisions for subsequent orders. The key point here is that market characteristics (e.g. volume, spreads and volatility) continually change. Therefore, measurement of execution quality should be continuous and part of the workflow of order execution in order that decisions on algo selection and venue are based on current market conditions and the extent to which they deviate from historical market conditions for similar orders.

The essential component required for such on-going analysis and decision-making is recording faithfully, in real-time, the full depth of the order book for each liquidity venue to which the firm is connected. With multiple liquidity providers (LPs), this is the proverbial “drinking from the firehose” which requires the ability for the trading system to ingest quote updates and trades at rates measured in hundreds of thousands per second. A key aspect here is that this quote recording system is plugged into the trading firm’s production trading infrastructure. As such, the latencies and infrastructure implicit in the trading firm’s particular set-up are baked into the historical time-series thus recorded, and it is more straightforward to intertwine orders, executions and the state of the order book at the time of each order and execution.

Once this recording system is part of the trading infrastructure, a time-series of trading-firm specific quotes and trades is automatically created and maintained. Again, by being part of the production trading infrastructure, this time-series production receives the necessary care and attention befitting such a valuable resource. This systematically maintained time-series database (or better, ‘knowledge base’) is the cornerstone of implementing a systematic approach to improving FX execution quality by appropriate algo and venue selection.

Of course, the lack of a central tape or official close prices as in equities and futures markets begs the question as to the relevant benchmark for measuring FX execution quality. But, when a firm is using TCA as part of a solution to improve execution quality (versus box-ticking), then the benchmark can and should be specific to the objectives of that firm’s FX trading (e.g. hedging versus intra-day trading).

Because of the fragmented nature of liquidity in the FX market, improving FX order execution is not just about how to execute (i.e. which execution algo to use). FX execution analysis has to operate across venues, so where to execute is also critical. This is preferably undertaken in real-time by a smart order routing (SOR) algo. SOR algos determine venue selection typically by looking at the best bid or offer simultaneously provided by each of the connected LPs. Depending on the time of day, currency pair, order size, required aggressiveness, etc, the SOR algo may send child orders to multiple LPs and may use multiple levels of liquidity. The intelligent SOR algo accounts for the historical fill ratio and rejections. Even if a venue offers the current best price, it may be more risky to route the flow to this venue if historically this location has a high rejection rate. Thus, it is essential to SOR operation to have access to the time-series of market data, orders and actual executions in order to calculate fill and rejection profiles for each liquidity venue. The calibration of the SOR algo should be continually evaluated. This is done by back-testing candidate SOR algos (including different parameterisation of the ‘same’ algo) against the firm’s own knowledge base of market data, orders and executions.

In summary, we need to combine traditional FX TCA with liquidity analysis across multiple venues and provide pre-trade analytics on current and historical data to provide traders with relevant and current analytics to optimise execution quality for the next order.

Related Content:

Order Execution Resources
AlgoCompass

This blog post by Stuart Farr originally appeared on the Best Execution website.

Recently, Deltix was spotlighted in two industry publications – CTA Intelligence and e-FOREX. Below are synopses of the two articles and links to view the original articles (no subscription or login needed).

DIY Execution Analysis

The CTA Intelligence Special Report: Managed Futures 2016 included an article by Stuart Farr titled “DIY Execution Analysis.” In the article, Stuart talks about the high cost and potentially undesirable characteristics of purchasing historical market data versus recording your own market data. Unless you’re running a start-up firm, it’s better to record real time market data for use in back-testing and execution analysis (or TCA).

The article discusses multiple advantages of recording your own market data. For example:

• You can capture all of the latencies and idiosyncrasies inherent in your own infrastructure and market data feeds. This is most important for strategies or analysis (like execution quality) that require tick and market depth data.
• If you’re doing forensic execution analysis, it is doubly necessary as the time-stamping of orders, executions and market data needs to be fully sequenced. Such forensic execution analysis may not be required on a continuous basis, but is essential when there is an unexpected or unexplained change in execution quality.
• For orders executed algorithmically, ongoing analysis is useful to track performance of the algos. Analysis can provide comfort or alert to unacceptable changes in execution quality measured in both ticks and dollar value. Time-series analysis of executions can be used to look for patterns of over/under performance and improve execution algo selection and parameterization.
• A practical key to success is to demonstrably and frequently improve trading performance by fine-tuning execution. There is nothing like showing real dollar improvement to keep researchers, technologists and traders motivated and focused.

Stuart goes into detail about each of these points. You can access the article here (no registration or login required).

Strategy Back Testing Platforms

e-FOREX published an article, “Sharpening the tools of the trade: New Developments with FX Strategy Back Testing Platforms”, that features comments from Ilya Gorelik on the topic of back testing foreign exchange (FX) trading strategies. This article talks about the challenges in signal generation and back testing in FX and how new technologies (e.g. cloud) can assist.

The subject of the e-FOREX article is back-testing only. As Deltix clients already appreciate, it is extremely important to be able to deploy a backtested trading strategy “as is” with no re-coding. This helps to protect the fidelity of actual results versus back-tested results. It also reduces time-to-market and the overall cost of the strategy development and deployment process. Deltix is the only institutional grade platform that provides this full, research-to-live-deployment capability.

You can read this article here (no registration or login required).

The Deltix Product Suite provides end-to-end support of all phases of the alpha discovery process, including data collection and aggregation, model development, back-testing, simulation and deployment to production. Learn more here.

Corporate Reputation Data

The Corporate Reputation Index (CRI), developed by Seldonix and distributed in partnership with Social Market Analytics, allows traders to identify and compare over- and under-priced stocks that are affected by a company’s actions. Corporate Reputation is the controllable portion of a company’s stock price, considered to be an intrinsic driver of company performance, and a major contributor to daily stock price fluctuation.

The Corporate Reputation Index is the first daily business metric for monitoring corporate reputation. Centered on “0”, companies with negative values are deemed undervalued relative to market while those with positive values are overvalued.

The CRI is created after the close of markets and is currently available before markets open the next day at any time after 9:00 PM Eastern US Time the night before.

Each data record features a set of fields:

• The timestamp associated with each field is the market close date of the data used;
• Symbol;
• CRI

Basis of Research using Seldonix data

The purpose of the research described in this paper is to determine if there are opportunities to generate alpha in US equities traded on the NYSE and NASDAQ using Seldonix data as a basis for daily market movement prediction (the holding period is about 6.5 hours).

We show how with the use of simple technical indicators and the popular conception of basket trading, we can exploit the CRI to generate excess returns. An evaluation of the CRI was conducted in the forecasting of stock price. The results of the testing showed a 1 – 8 day lag in the prediction of stock price, depending on the stock. Therefore our purpose is to determine the most appropriate time when the CRI generates the most suitable values for opening a position. As the CRI values change every day and are different for each ticker, we need to construct an approach which will successfully encompass both equally important collations. To identify these situations, we need to:

• Compare short-term values to long-term;
• Compare values within the considered universe of tickers.

The full paper can be accessed here.

Results

This approach gives the following results for S&P 100 tested on the period from 1/5/2010 to 8/18/2014:

Parameter	All Trades	Long Trades	Short Trades
Net Profit/Loss	21,182.82	13,984.97	7,197.85
Total Profit	113,708.71	59,129.80	54,578.91
Total Loss	-92,525.89	-45,144.83	-47,381.06
Cumulated Profit %	21.18 %	13.98 %	7.20 %
Max Drawdown	-2,590.71	-2,774.62	-2,265.55
Max Drawdown %	-2.41 %	-2.66 %	-2.18 %
Max Drawdown Date	9/2/2011	9/30/2011	9/15/2011
Return/Drawdown Ratio	8.18	5.04	3.18
Drawdown Days %	80.96 %	80.36 %	88.29 %
Max Drawdown Duration	119	198	188
CAGR	4.40 %	2.97 %	1.57 %
Sharpe Ratio	1.73	1.37	0.73
Annualized Volatility	2.54	2.16	2.15
Sortino Ratio	2.86	2.15	1.13
UPI	0.39	0.27	0.13
Information Ratio	1.71	1.37	0.72
Optimal f	68.32	63.51	33.96
All Trades #	2308	1166	1142
Profitable Trades Ratio	0.53	0.54	0.52
Winning Trades #	1229	634	595
Losing Trades #	1079	532	547
Average Trade	9.18	11.99	6.30
Average Winning Trade	92.52	93.26	91.73
Average Losing Trade	-85.75	-84.86	-86.62
Avg. Win/ Avg. Loss Ratio	1.08	1.10	1.06
Average Profit per Share	0.04	0.05	0.03
Max Consequent Winners	9	11	9
Max Consequent Losers	10	10	9

 

Rules for closing

The other idea is to exploit the CR trend not only for opening but for closing position. On the figure below you can see the example of position on AAPL chart, blue line on second pad indicates CRI values:

We use the Simple Moving Average (SMA) of the last CRI values (red line on the second pad) to determine whether the instrument gets over or underpriced. Using the model from the paragraph Over/Under Priced, we will close the position for the instrument according to following rule:

1. If CR > SMA (5, CR) (instrument is overpriced at time t, price is likely to return to suggested price at t+1) close Short position on day close;
2. If CR < SMA (5, CR) (instrument is underpriced at time t, price is likely to return to suggested price at t+1) close Long position on day close.

The same chart with this rule shows better results:

This approach gives the following results for S&P100 tested on the period from 1/5/2010 to 8/18/2014:

Parameter	All Trades	Long Trades	Short Trades
Net Profit/Loss	32,118.49	18,265.41	13,853.08
Total Profit	143,984.75	71,784.79	72,199.96
Total Loss	-111,866.26	-53,519.38	-58,346.88
Cumulated Profit %	32.12 %	18.27 %	13.85 %
Max Drawdown	-3,526.80	-4,508.75	-3,368.95
Max Drawdown %	-3.08 %	-4.22 %	-3.07 %
Max Drawdown Date	9/2/2011	9/30/2011	9/15/2011
Return/Drawdown Ratio	9.11	4.05	4.11
Drawdown Days %	80.45 %	81.83 %	88.63 %
Max Drawdown Duration	95	194	279
CAGR	6.44 %	3.83 %	2.95 %
Sharpe Ratio	1.97	1.31	0.94
Annualized Volatility	3.27	2.93	3.14
Sortino Ratio	3.40	2.06	1.55
UPI	0.58	0.26	0.19
Information Ratio	1.90	1.31	0.91
Optimal f	60.05	44.63	29.93
All Trades #	2190	1103	1087
Profitable Trades Ratio	0.53	0.55	0.51
Winning Trades #	1158	603	555
Losing Trades #	1032	500	532
Average Trade	14.67	16.56	12.74
Average Winning Trade	124.34	119.05	130.09
Average Losing Trade	-108.40	-107.04	-109.67
Avg. Win/ Avg. Loss Ratio	1.15	1.11	1.19
Average Profit per Share	0.06	0.07	0.05
Max Consequent Winners	13	16	16
Max Consequent Losers	11	9	15

 

Filter for stronger signal

One more idea is to filter a stronger signal when CR retains its sign from the previous day, which is a stronger signal indicating overpriced/ underpriced asset. On the chart below you can see two trades on MSFT asset, the second one is unprofitable, when the CR line changes its value from -1 to 1:

Filtering out such unstable situations gets the following results for the S&P 100 tested in the period from 1/5/2010 to 8/18/2014:

Parameter	All Trades	Long Trades	Short Trades
Net Profit/Loss	32,810.12	19,145.45	13,664.67
Total Profit	126,322.99	64,003.72	62,319.27
Total Loss	-93,512.87	-44,858.26	-48,654.61
Cumulated Profit %	32.81 %	19.15 %	13.66 %
Max Drawdown	-2,617.49	-4,409.37	-2,465.16
Max Drawdown %	-2.37 %	-4.18 %	-2.29 %
Max Drawdown Date	9/2/2011	9/30/2011	9/15/2011
Return/Drawdown Ratio	12.53	4.34	5.54
Drawdown Days %	78.38 %	80.02 %	86.74 %
Max Drawdown Duration	89	194	176
CAGR	6.56 %	4.00 %	2.91 %
Sharpe Ratio	2.16	1.49	1.03
Annualized Volatility	3.04	2.68	2.84
Sortino Ratio	3.82	2.45	1.70
UPI	0.67	0.26	0.24
Information Ratio	2.11	1.50	1.01
Optimal f	70.80	55.51	36.20
All Trades #	1839	937	902
Profitable Trades Ratio	0.53	0.55	0.51
Winning Trades #	972	512	460
Losing Trades #	867	425	442
Average Trade	17.84	20.43	15.15
Average Winning Trade	129.96	125.01	135.48
Average Losing Trade	-107.86	-105.55	-110.08
Avg. Win/ Avg. Loss Ratio	1.20	1.18	1.23
Average Profit per Share	0.07	0.08	0.06
Max Consequent Winners	12	12	11
Max Consequent Losers	11	10	16

 

Conclusion

We presented an approach that uses CR indices as a strong predictive factor of price directionality. We developed a trading strategy that implements an algorithm based on this approach. We also constructed two other versions, both based on CRI predictive power, that raise the performance of the main approach.

The main contribution of the paper is back-testing and comparison of the different versions of the strategy. On stocks in the S&P 100, back-testing shows that the first strategy has an average Information Ratio of 1.71 over the period 2010-2014, the second and third versions achieved Information Ratios of 1.90 and 2.11 respectively, which indicates that the CRI is predictive in the forecasting of stock returns.

By: The Deltix Quantitative Research Team

Introduction

This paper describes the implementation of an automated equity trading strategy based on aggregated company earnings estimates from independent, buy-side, and sell-side analysts, along with those of private investors and students. By sourcing estimates from a diverse community of individuals (“crowd-sourcing”), Estimize provides an alternative view of earnings expectations compared to traditional sell-side analysts.

We implemented the strategy in the Deltix QuantOffice research platform, a purpose-built C# development studio with embedded math, statistics and data libraries.

Estimize Data

The Estimize data used in this research is a set of earnings per share (EPS) estimates. Each data record refers to a certain EPS announcement which is made for a stock every quarter. The data record features a set of fields:

–        Instrument type;

–        Creation time;

–        Estimate_id;

–        Released_id;

–        EPS forecast;

–        Revenue forecast;

–        Date of the release.

 

A sample of the Estimize data as represented in Deltix QuantOffice is shown below.

 

Basis of Research using Estimize data

The purpose of the research described in this paper is to determine if there are opportunities to generate alpha in US equities, using Estimize data as a basis for overnight market movement prediction (the holding period is about 19 hours). We show how with simple mathematical techniques, we can identify pricing inefficiencies which we can exploit to generate excess returns. We take the theory of stock market overreaction as a basis for our model. A consequence of overreaction is the possible profitability of a contrarian strategy, that is, a strategy that exploits negative serial dependence in asset returns. The defining characteristic of a contrarian strategy is the purchase of securities that have performed poorly in the past and the sale of securities that have performed well. Such “selling the winners” and “buying the losers” may earn positive returns because current losers are likely to become future winners and current winners are likely to become future losers when stock returns are negatively autocorrelated. Therefore, it may be said that an implication of stock market overreaction is positive expected profits from a contrarian investment strategy.

This research analyzes how Estimize forecasts change over the period preceding the announcement by a company of its earnings (reporting period). The directionality in estimates can serve as an indicator for “winners” and “losers”. We determine if the short term estimates are more optimistic or pessimistic than the long term ones to predict the change of price after the announcement.

As a start, let us determine which estimates should be considered as short term. To do this, the average number of estimates per reporting period for a ticker universe of the S&P100 is calculated. The results are shown below.

Ticker	Number	Ticker	Number	Ticker	Number
AAPL	186.57	MON	10.43	BMY	4.13
GOOG	70.62	HON	10.33	FCX	4.00
AMZN	53.10	JNJ	9.64	APC	3.88
NKE	40.22	CVS	9.23	MET	3.80
SBUX	33.00	WMT	9.15	UNP	3.78
EBAY	32.85	PG	9.00	WAG	3.67
INTC	30.46	FDX	8.50	RTN	3.43
DIS	28.00	CL	7.50	UPS	3.43
JPM	27.56	UNH	7.50	LLY	3.36
CSCO	25.91	AIG	7.44	AEP	3.33
IBM	24.85	MRK	7.25	DOW	3.00
F	24.45	GILD	7.11	NOV	2.75
MSFT	22.92	ABT	6.78	NSC	2.75
V	22.64	HAL	6.67	MDT	2.67
ORCL	21.30	AMGN	6.63	SO	2.67
QCOM	19.83	BAX	6.57	MMM	2.63
HPQ	19.73	EMR	6.57	MDLZ	2.33
MA	18.38	PEP	6.50	UTX	2.33
HD	17.36	BA	6.44	DVN	2.25
DELL	16.71	T	6.44	EXC	2.25
EMC	16.25	LMT	6.29	COF	2.09
WFC	16.00	CVX	6.09	COP	2.00
LOW	15.88	MS	5.83	MO	2.00
GS	15.62	PFE	5.50	OXY	2.00
TGT	14.56	VZ	5.13	BK	1.88
COST	13.82	TWX	5.00	APA	1.75
TXN	13.77	SLB	4.78	PM	1.50
BAC	13.58	AXP	4.69	WMB	1.00
C	13.00	USB	4.50	ALL	0.00
MCD	13.00	XOM	4.40	BRK.B	0.00
CAT	11.31	DD	4.25	SPG	0.00
GE	11.11	ACN	4.22
KO	10.64	GD	4.20

 

The mean value is 12.5 estimates per reporting period. To designate estimates as a short term, we need to balance between having too large a number of estimates (which may increase imprecision) and too small a number (which may be unrepresentative). Therefore, we expect that the boundary between these short term and long term estimates should be a particular quintile, q, of estimates:

In our implementation q= 0.2, which means that for 100 estimates, we take the mean value of the last 20 as the short term forecast.

To measure the predictive power of this approach we need to collect the estimates prior to the last reporting date and calculate the mean estimate which is our long term forecast:

The last estimate in this summation is at the day close prior to the report.

Similarly, the short term forecast is:

Trading Hypothesis

Our hypothesis is that relatively higher optimism in the short term forecast compared to the long-term forecast usually leads to negative price movement after the actual earnings announcement. In this case, the announcement will probably miss the short term expectation, and the stock price will fall. On the other hand, where long term optimism is followed by short term, pre-announcement, pessimism then this predicts positive stock price movement.

Trading Algorithm

To implement this hypothesis as an automated trading strategy we need to:

1)     Collect the earnings estimates until the day close prior to the earnings announcement and calculate the mean estimate:

meanEPS = sumEPS/EPScount;

 

2)     Calculate the mean of last q values:

k  = Floor(EPScount*q);

meanEPS_q = sumEPS_k/k;

 

3)     Calculate the difference between long and short term forecast, i.e. the difference between the mean Estimize consensus pre-report and mean Estimize consensus of the last 20% values:

 

delta = meanEPS – meanEPS_q;

 

4)     Open position prior to the EPS announcement in accordance with ∆.

 if (delta < 0) goShort; else goLong;

 

5)     Close position on next day open.

In the strategy, we open the position at 15:59 and close the position at 09:35 next day. Through testing, we determined that opening position at the closing auction price has no significant impact on performance.

 

Market Neutrality

We tested strategy performance using dollar (notional) and beta (market) neutral approaches with the goal of creating a portfolio uncorrelated with the movement of the market, i.e. to determine that this Estimize-based strategy generates true alpha.

 

1)     Dollar-neutral:

The portfolio is hedged using the S&P 500 index via the SPY. We calculate the dollar amount of all opened positions at 15:59 and open the opposite position in SPY. If the dollar amounts invested in each of the stocks are  , then we invest in SPY:  

 

 2)     Beta-neutral:

We performed a regression of stock returns on the SPY return over a period of 250 days:

 

Where denotes the daily return of a stock i or SPY index respectively, R = log(CurrentDayClose/prevDayClose). We used Ordinary Least Squares method to estimate betas for each stock.

If the dollar amounts invested in each of the stocks are , then the dollar amount invested in SPY is such that:

 

which keeps the overall portfolio beta-neutral.

These variations of the algorithm are available via strategy input parameters, in QuantOffice.

Consolidated Report

The strategy was run on securities:

1) S&P100;

2) 100 companies from S&P500 with the greatest turnover;

3) 200 companies from S&P500 with the greatest turnover.

Back-testing period: from 12/26/2011 to 12/31/2013.

 

	S&P100			100 companies from S&P500			200 companies from S&P500
Strategy	Stocks	Dollar neutral	Beta neutral	Stocks	Dollar neutral	Beta neutral	Stocks	Dollar neutral	Beta neutral
Net Profit/Loss	22,903.79	18,235.72	17,099.26	28,831.43	24,290.54	21,602.58	42,221.01	34,195.22	28,497.21
Total Profit	88,561.18	92,850.19	88,344.25	121,216.22	125,820.11	121,110.75	211,465.74	219,339.25	207,103.44
Total Loss	-65,657.39	-74,614.47	-71,244.99	-92,384.79	-101,529.57	-99,508.17	-169,244.74	-185,144.03	-178,606.22
Cumulated Profit %	22.90 %	18.24 %	17.10 %	28.83 %	24.29 %	21.60 %	42.22 %	34.20 %	28.50 %
Max Drawdown	-4,050.98	-4,064.70	-4,130.32	-6,097.02	-6,023.01	-5,886.81	-6,847.01	-6,567.76	-6,992.91
Max Drawdown %	-3.44 %	-3.54 %	-3.62 %	-4.80 %	-4.84 %	-4.82 %	-4.87 %	-4.73 %	-5.23 %
Max Drawdown Time	1/28/2013	1/28/2013	1/28/2013	1/23/2013	1/23/2013	1/23/2013	3/19/2013	11/27/2013	11/29/2013
Return/Drawdown Ratio	5.65	4.49	4.14	4.73	4.03	3.67	6.17	5.21	4.08
Drawdown Days %	69.17 %	72.33 %	48.17 %	72.33 %	72.92 %	49.12 %	80.43 %	82.61 %	55.37 %
Max Drawdown Duration	81	81	81	137	139	139	78	133	133
CAGR	11.18 %	8.99 %	5.71 %	13.90 %	11.82 %	7.12 %	19.84 %	16.31 %	9.27 %
Sharpe Ratio	2.08	1.72	1.36	2.12	1.83	1.33	2.19	1.88	1.33
Sortino Ratio	5.38	5.22	4.20	6.56	6.47	5.34	9.06	8.66	6.99
UPI	3.43	2.74	2.16	3.77	3.13	2.26	3.90	3.14	2.14
Annualized Volatility	0.72	0.54	0.40	0.65	0.46	0.32	0.64	0.46	0.30
Information Ratio	1.93	1.62	1.29	1.88	1.63	1.22	1.87	1.62	1.18
Optimal f	38.65	33.01	32.41	32.33	28.21	24.98	24.20	21.77	19.01
All Trades #	603	784	771	657	838	823	1148	1389	1344
Profitable Trades Ratio	0.54	0.52	0.52	0.54	0.52	0.52	0.53	0.52	0.52
Winning Trades #	328	409	401	355	433	425	607	718	693
Losing Trades #	275	375	370	302	405	398	541	671	651
Average Trade	37.98	23.26	22.18	43.88	28.99	26.25	36.78	24.62	21.20
Average Winning Trade	270.00	227.02	220.31	341.45	290.58	284.97	348.38	305.49	298.85
Average Losing Trade	-238.75	-198.97	-192.55	-305.91	-250.69	-250.02	-312.84	-275.92	-274.36
Avg. Win/ Avg. Loss Ratio	1.13	1.14	1.14	1.12	1.16	1.14	1.11	1.11	1.09
Average Profit per Share	0.17	0.12	0.11	0.19	0.14	0.13	0.15	0.11	0.09
Max Conseq. Winners	8	10	10	7	8	7	10	8	7
Max Conseq. Losers	6	6	7	9	7	7	8	8	8

 

Conclusion

We presented an approach that measures the relative optimism or pessimism of short term versus long term Estimize estimates as a strong predictive factor of the price directionality after EPS announcements. We developed a trading strategy that implements an algorithm based on this approach. We also constructed dollar-neutral and beta-neutral portfolios to demonstrate that the achieved effect is uncorrelated with the market.

The main contribution of the paper is back-testing and comparison of the unhedged strategy with dollar-neutral and beta-neutral ones. On stocks in the S&P100, back-testing shows that the unhedged strategy has an Information ratio of 1.93 over the period 2012-2013, with a stronger performance in 2012. Dollar-neutral and beta-neutral strategies achieved Information Ratios of 1.62 and 1.29 respectively, which indicates that the strategy does generate alpha.

 

 

 

Appendix A: Detailed Reports

Only S&P100 stocks:

Net Profit/Loss	22,903.79	13,834.76	9,069.03
Total Profit	88,561.18	58,373.43	30,187.75
Total Loss	-65,657.39	-44,538.67	-21,118.72
Cumulated Profit %	22.90 %	13.83 %	9.07 %
Max Drawdown	-4,050.98	-2,535.09	-2,528.59
Max Drawdown %	-3.44 %	-2.28 %	-2.32 %
Max Drawdown Date	1/28/2013	1/24/2013	7/31/2013
Return/Drawdown Ratio	5.65	5.46	3.59
Drawdown Days %	69.17 %	78.26 %	73.91 %
Max Drawdown Duration	81	83	137
CAGR	11.18 %	6.88 %	4.56 %
Sharpe Ratio	2.08	1.49	1.33
Annualized Volatility	5.38	4.62	3.43
Sortino Ratio	3.43	2.43	2.32
UPI	0.72	0.52	0.34
Information Ratio	1.93	1.42	1.29
Optimal f	38.65	32.26	38.88
All Trades #	603	441	162
Profitable Trades Ratio	0.54	0.54	0.56
Winning Trades #	328	238	90
Losing Trades #	275	203	72
Average Trade	37.98	31.37	55.98
Average Winning Trade	270.00	245.27	335.42
Average Losing Trade	-238.75	-219.40	-293.32
Avg. Win/ Avg. Loss Ratio	1.13	1.12	1.14
Average Profit per Share	0.17	0.15	0.23
Max Conseq. Winners	8	7	7
Max Conseq. Losers	6	6	5

Hedged by SPY S&P100 stocks:

 

Net Profit/Loss	18,235.72	13,519.25	4,716.47
Total Profit	92,850.19	58,917.72	33,932.47
Total Loss	-74,614.47	-45,398.47	-29,216.00
Cumulated Profit %	18.24 %	13.52 %	4.72 %
Max Drawdown	-4,064.70	-2,553.26	-3,319.65
Max Drawdown %	-3.54 %	-2.30 %	-3.13 %
Max Drawdown Date	1/28/2013	1/24/2013	8/1/2013
Return/Drawdown Ratio	4.49	5.29	1.42
Drawdown Days %	72.33 %	83.79 %	90.91 %
Max Drawdown Duration	81	107	169
CAGR	8.99 %	6.73 %	2.40 %
Sharpe Ratio	1.72	1.45	0.63
Annualized Volatility	5.22	4.63	3.80
Sortino Ratio	2.74	2.37	0.99
UPI	0.54	0.50	0.12
Information Ratio	1.62	1.39	0.62
Optimal f	33.01	31.44	16.63
All Trades #	784	483	301
Profitable Trades Ratio	0.52	0.53	0.50
Winning Trades #	409	257	152
Losing Trades #	375	226	149
Average Trade	23.26	27.99	15.67
Average Winning Trade	227.02	229.25	223.24
Average Losing Trade	-198.97	-200.88	-196.08
Avg. Win/ Avg. Loss Ratio	1.14	1.14	1.14
Average Profit per Share	0.12	0.14	0.08
Max Conseq. Winners	10	7	6
Max Conseq. Losers	6	6	8

Beta neutral S&P100 stocks:

Net Profit/Loss	17,099.26	13,140.04	3,959.22
Total Profit	88,344.25	55,753.20	32,591.05
Total Loss	-71,244.99	-42,613.16	-28,631.83
Cumulated Profit %	17.10 %	13.14 %	3.96 %
Max Drawdown	-4,130.32	-2,589.82	-3,407.28
Max Drawdown %	-3.62 %	-2.34 %	-3.24 %
Max Drawdown Date	1/28/2013	1/24/2013	8/1/2013
Return/Drawdown Ratio	4.14	5.07	1.16
Drawdown Days %	48.17 %	54.53 %	60.76 %
Max Drawdown Duration	81	129	169
CAGR	5.71 %	4.44 %	1.38 %
Sharpe Ratio	1.36	1.20	0.44
Annualized Volatility	4.20	3.71	3.13
Sortino Ratio	2.16	1.93	0.67
UPI	0.40	0.39	0.08
Information Ratio	1.29	1.15	0.44
Optimal f	32.41	32.23	14.08
All Trades #	771	469	302
Profitable Trades Ratio	0.52	0.54	0.49
Winning Trades #	401	253	148
Losing Trades #	370	216	154
Average Trade	22.18	28.02	13.11
Average Winning Trade	220.31	220.37	220.21
Average Losing Trade	-192.55	-197.28	-185.92
Avg. Win/ Avg. Loss Ratio	1.14	1.12	1.18
Average Profit per Share	0.11	0.15	0.07
Max Conseq. Winners	10	7	6
Max Conseq. Losers	7	6	6

 

Only 100 stocks from S&P500:

Net Profit/Loss	28,831.43	23,711.21	5,120.22
Total Profit	121,216.22	80,130.00	41,086.22
Total Loss	-92,384.79	-56,418.79	-35,966.00
Cumulated Profit %	28.83 %	23.71 %	5.12 %
Max Drawdown	-6,097.02	-3,303.87	-5,767.64
Max Drawdown %	-4.80 %	-2.60 %	-5.24 %
Max Drawdown Date	1/23/2013	12/20/2013	5/21/2013
Return/Drawdown Ratio	4.73	7.18	0.89
Drawdown Days %	72.33 %	65.42 %	78.46 %
Max Drawdown Duration	137	97	301
CAGR	13.90 %	11.55 %	2.60 %
Sharpe Ratio	2.12	2.17	0.57
Annualized Volatility	6.56	5.32	4.59
Sortino Ratio	3.77	4.18	0.87
UPI	0.65	0.83	0.05
Information Ratio	1.88	1.98	0.54
Optimal f	32.33	40.79	12.33
All Trades #	657	451	206
Profitable Trades Ratio	0.54	0.55	0.52
Winning Trades #	355	247	108
Losing Trades #	302	204	98
Average Trade	43.88	52.57	24.86
Average Winning Trade	341.45	324.41	380.43
Average Losing Trade	-305.91	-276.56	-367.00
Avg. Win/ Avg. Loss Ratio	1.12	1.17	1.04
Average Profit per Share	0.19	0.23	0.11
Max Conseq. Winners	7	8	7
Max Conseq. Losers	9	10	5

Hedged by SPY 100 stocks:

 

Net Profit/Loss	24,290.54	23,623.01	667.54
Total Profit	125,820.11	81,115.72	44,704.39
Total Loss	-101,529.57	-57,492.71	-44,036.86
Cumulated Profit %	24.29 %	23.62 %	.67 %
Max Drawdown	-6,023.01	-3,452.13	-7,171.01
Max Drawdown %	-4.84 %	-2.72 %	-6.68 %
Max Drawdown Date	1/23/2013	12/20/2013	10/28/2013
Return/Drawdown Ratio	4.03	6.84	0.09
Drawdown Days %	72.92 %	75.30 %	86.56 %
Max Drawdown Duration	139	100	301
CAGR	11.82 %	11.51 %	.34 %
Sharpe Ratio	1.83	2.16	0.07
Annualized Volatility	6.47	5.32	4.93
Sortino Ratio	3.13	4.17	0.10
UPI	0.46	0.79	0.01
Information Ratio	1.63	1.97	0.07
Optimal f	28.21	40.61	1.41
All Trades #	838	503	335
Profitable Trades Ratio	0.52	0.54	0.49
Winning Trades #	433	270	163
Losing Trades #	405	233	172
Average Trade	28.99	46.96	1.99
Average Winning Trade	290.58	300.43	274.26
Average Losing Trade	-250.69	-246.75	-256.03
Avg. Win/ Avg. Loss Ratio	1.16	1.22	1.07
Average Profit per Share	0.14	0.22	0.01
Max Conseq. Winners	8	9	6
Max Conseq. Losers	7	8	6

Beta neutral 100 stocks

 

Net Profit/Loss	21,602.58	22,202.86	-600.28
Total Profit	121,110.75	77,638.42	43,472.33
Total Loss	-99,508.17	-55,435.57	-44,072.61
Cumulated Profit %	21.60 %	22.20 %	-.60 %
Max Drawdown	-5,886.81	-3,471.30	-7,440.96
Max Drawdown %	-4.82 %	-2.76 %	-6.99 %
Max Drawdown Date	1/23/2013	12/20/2013	10/28/2013
Return/Drawdown Ratio	3.67	6.40	-0.08
Drawdown Days %	49.12 %	50.88 %	59.00 %
Max Drawdown Duration	139	100	301
CAGR	7.12 %	7.31 %	-.21 %
Sharpe Ratio	1.33	1.68	-0.05
Annualized Volatility	5.34	4.35	4.12
Sortino Ratio	2.26	3.21	-0.07
UPI	0.32	0.59	0.00
Information Ratio	1.22	1.56	-0.05
Optimal f	24.98	38.58	-1.25
All Trades #	823	490	333
Profitable Trades Ratio	0.52	0.54	0.47
Winning Trades #	425	267	158
Losing Trades #	398	223	175
Average Trade	26.25	45.31	-1.80
Average Winning Trade	284.97	290.78	275.14
Average Losing Trade	-250.02	-248.59	-251.84
Avg. Win/ Avg. Loss Ratio	1.14	1.17	1.09
Average Profit per Share	0.13	0.22	-0.01
Max Conseq. Winners	7	9	6
Max Conseq. Losers	7	8	7

Only 200 stocks from S&P500:

 

Net Profit/Loss	42,221.01	35,965.79	6,255.22
Total Profit	211,465.74	148,821.21	62,644.53
Total Loss	-169,244.74	-112,855.42	-56,389.31
Cumulated Profit %	42.22 %	35.97 %	6.26 %
Max Drawdown	-6,847.01	-5,999.18	-6,399.88
Max Drawdown %	-4.87 %	-4.60 %	-5.73 %
Max Drawdown Date	3/19/2013	2/20/2013	5/21/2013
Return/Drawdown Ratio	6.17	6.00	0.98
Drawdown Days %	80.43 %	82.21 %	87.15 %
Max Drawdown Duration	78	109	299
CAGR	19.84 %	17.10 %	3.17 %
Sharpe Ratio	2.19	2.09	0.55
Annualized Volatility	9.06	8.18	5.72
Sortino Ratio	3.90	3.90	0.87
UPI	0.64	0.62	0.07
Information Ratio	1.87	1.85	0.53
Optimal f	24.20	25.58	9.69
All Trades #	1148	841	307
Profitable Trades Ratio	0.53	0.54	0.50
Winning Trades #	607	453	154
Losing Trades #	541	388	153
Average Trade	36.78	42.77	20.38
Average Winning Trade	348.38	328.52	406.78
Average Losing Trade	-312.84	-290.86	-368.56
Avg. Win/ Avg. Loss Ratio	1.11	1.13	1.10
Average Profit per Share	0.15	0.18	0.09
Max Conseq. Winners	10	9	5
Max Conseq. Losers	8	8	5

Hedged by SPY 200 stocks:

 

Net Profit/Loss	34,195.22	35,888.21	-1,692.99
Total Profit	219,339.25	149,819.75	69,519.49
Total Loss	-185,144.03	-113,931.54	-71,212.48
Cumulated Profit %	34.20 %	35.89 %	-1.69 %
Max Drawdown	-6,567.76	-5,997.23	-9,581.66
Max Drawdown %	-4.73 %	-4.60 %	-8.91 %
Max Drawdown Date	11/27/2013	2/20/2013	11/8/2013
Return/Drawdown Ratio	5.21	5.98	-0.18
Drawdown Days %	82.61 %	83.99 %	92.29 %
Max Drawdown Duration	133	109	299
CAGR	16.31 %	17.07 %	-.87 %
Sharpe Ratio	1.88	2.09	-0.14
Annualized Volatility	8.66	8.18	6.47
Sortino Ratio	3.14	3.89	-0.19
UPI	0.46	0.62	-0.01
Information Ratio	1.62	1.85	-0.13
Optimal f	21.77	25.53	-2.09
All Trades #	1389	892	497
Profitable Trades Ratio	0.52	0.54	0.48
Winning Trades #	718	479	239
Losing Trades #	671	413	258
Average Trade	24.62	40.23	-3.41
Average Winning Trade	305.49	312.78	290.88
Average Losing Trade	-275.92	-275.86	-276.02
Avg. Win/ Avg. Loss Ratio	1.11	1.13	1.05
Average Profit per Share	0.11	0.17	-0.02
Max Conseq. Winners	8	9	7
Max Conseq. Losers	8	8	6

Beta neutral 200 stocks:

 

Net Profit/Loss	28,497.21	33,793.76	-5,296.55
Total Profit	207,103.44	141,269.44	65,834.00
Total Loss	-178,606.22	-107,475.67	-71,130.55
Cumulated Profit %	28.50 %	33.79 %	-5.30 %
Max Drawdown	-6,992.91	-5,998.04	-10,306.36
Max Drawdown %	-5.23 %	-4.68 %	-9.85 %
Max Drawdown Date	11/29/2013	2/20/2013	11/8/2013
Return/Drawdown Ratio	4.08	5.63	-0.51
Drawdown Days %	55.37 %	56.46 %	62.86 %
Max Drawdown Duration	133	109	299
CAGR	9.27 %	10.85 %	-1.91 %
Sharpe Ratio	1.33	1.64	-0.34
Annualized Volatility	6.99	6.61	5.55
Sortino Ratio	2.14	3.07	-0.47
UPI	0.30	0.47	-0.03
Information Ratio	1.18	1.48	-0.34
Optimal f	19.01	24.81	-6.19
All Trades #	1344	853	491
Profitable Trades Ratio	0.52	0.54	0.47
Winning Trades #	693	460	233
Losing Trades #	651	393	258
Average Trade	21.20	39.62	-10.79
Average Winning Trade	298.85	307.11	282.55
Average Losing Trade	-274.36	-273.47	-275.70
Avg. Win/ Avg. Loss Ratio	1.09	1.12	1.02
Average Profit per Share	0.09	0.17	-0.05
Max Conseq. Winners	7	9	7
Max Conseq. Losers	8	8	6