crowdsource earnings predictions and the quantopian research platform

41
Estimize and Quantopian Crowdsource Earnings Predictions and the Research Platform Quantopian Seong Lee (about )

Upload: quantopian

Post on 14-Jun-2015

791 views

Category:

Economy & Finance


4 download

DESCRIPTION

In this presentation, we will show you examples on how to incorporate multiple sources of earning predictions into your algorithms and why the sources matter. You will also get a sneak peek of our new beta research environment - where you can use IPython notebooks to analyze curated datasets, algorithms, and backtest results.

TRANSCRIPT

Page 1: Crowdsource Earnings Predictions and the Quantopian Research Platform

Estimize and QuantopianCrowdsource Earnings Predictions and the

Research Platform

QuantopianSeong Lee (about)

Page 2: Crowdsource Earnings Predictions and the Quantopian Research Platform

What are we talking about?

● EPS, Wall Street Consensus, Earnings Surprise

● Estimize, the crowdsourced financials aggregator

● Cover Estimize’s whitepaper and replicate the results in the Quantopian Research Platform

● Develop a simple trading strategy

Page 3: Crowdsource Earnings Predictions and the Quantopian Research Platform

By the end of this talk you’ll have

● Tasted the potential for crowdsourced financial data

● A sense of how the Quantopian Research Platform works and how to run parameter optimization

Page 4: Crowdsource Earnings Predictions and the Quantopian Research Platform

What you need to know

● EPS: Earnings Per Share o (Net Income - Dividends)/(Shares Outstanding)o An important factor in predicting a company’s

performanceo I.E. Did it increase from last quarter?

Page 5: Crowdsource Earnings Predictions and the Quantopian Research Platform

What you need to know

● Wall Street Consensuso Aggregated consensus of Wall Street Analysts

(mostly sell-side)o Sell-side: Investment Banks, FAs, Brokerso Buy-side: Hedge funds, mutual funds, pension

funds, VCs, Prop firms, PEso IBES (Institutional Brokers’ Estimate System)

Page 6: Crowdsource Earnings Predictions and the Quantopian Research Platform

What you need to know

● Earnings Surpriseo When announcements do better than or worse than

earnings estimateso Look at Apple Q2:

Actual: 1.66 Consensus: 1.46 Surprise: 13.7%

Page 7: Crowdsource Earnings Predictions and the Quantopian Research Platform

Summary

● EPSo Earnings Per Share

● Wall Street Consensuso Average of all Wall Street Estimates

● Earnings Surpriseo When earnings announcements differ from

expectations

Page 8: Crowdsource Earnings Predictions and the Quantopian Research Platform

Who is Estimize?

● Estimize is a financial estimates aggregatoro “Independent, buy-side, and sell-side analysts as

well as private investors and students”o EPS and Revenue estimates side-by-side with Wall

Street Consensus estimates

Page 9: Crowdsource Earnings Predictions and the Quantopian Research Platform

Revisiting Apple Q2, 2014

Page 10: Crowdsource Earnings Predictions and the Quantopian Research Platform

Crowdsourced Estimates

● What makes them different?o Diversity of contributors

Only 7% of total participants are sell-side analysts

Page 11: Crowdsource Earnings Predictions and the Quantopian Research Platform

Looking at the whitepaper claims

● Released September 24, 2013● Claim #1: “More accurate 65% of the time

when there are 20 or more contributors”● Claim #2: “Average absolute error of the

Estimize consensus is smaller than the Wall Street Consensus by 12 bp when contributors are greater than 20”

Page 12: Crowdsource Earnings Predictions and the Quantopian Research Platform

The Toolbox

Page 13: Crowdsource Earnings Predictions and the Quantopian Research Platform

Claim #1: “65% more accurate”

● Did Estimize land higher than Wall Street when it was a positive surprise?

● Did Estimize land lower than Wall Street when it was a negative surprise?

Page 14: Crowdsource Earnings Predictions and the Quantopian Research Platform

Revisiting Apple Q2, 2014

Page 15: Crowdsource Earnings Predictions and the Quantopian Research Platform

Implementation

● Compare the number of times that Estimize correctly guessed direction

● Data was preprocessed so feel free to reach out if you want steps

Page 16: Crowdsource Earnings Predictions and the Quantopian Research Platform

Wrangling our DataFrame

● We have 13,000 rows, each with this data

● So for each row, see if estimize predicted direction correctly

Page 17: Crowdsource Earnings Predictions and the Quantopian Research Platform

Loading our Data

Page 18: Crowdsource Earnings Predictions and the Quantopian Research Platform

Prediction Direction

Page 19: Crowdsource Earnings Predictions and the Quantopian Research Platform

Plot the results

Page 20: Crowdsource Earnings Predictions and the Quantopian Research Platform

What about the number of participants?● Each announcement can have as little as 1

participants or more than 167

● Apple Q2, 2014:

Page 21: Crowdsource Earnings Predictions and the Quantopian Research Platform

Plotting against N participants

Page 22: Crowdsource Earnings Predictions and the Quantopian Research Platform

Plotting against N participants

Page 23: Crowdsource Earnings Predictions and the Quantopian Research Platform

Summary

● Positive correlation between number of participants and the accuracy rate of Estimize versus the Wall Street Consensus

● Accuracy > 65% when N > 20

Page 24: Crowdsource Earnings Predictions and the Quantopian Research Platform

Claim #2: Lower absolute error

● We’re going to look at the relative delta of each estimate instead

● Steps:1. Wrangle our DataFrame/spreadsheet (add a

column)2. Plot the results against N participants

Page 25: Crowdsource Earnings Predictions and the Quantopian Research Platform

Wrangling our data

Page 26: Crowdsource Earnings Predictions and the Quantopian Research Platform

Graphing the score: Code

Page 27: Crowdsource Earnings Predictions and the Quantopian Research Platform

Graphing the score: Plot

Page 28: Crowdsource Earnings Predictions and the Quantopian Research Platform

Conclusions

● Claim #1: Positive relationship between accuracy and number of participantso Matches up

● Claim #2: Lower relative error as number of participants increaseso Matches up

● So how do we use this data?

Page 29: Crowdsource Earnings Predictions and the Quantopian Research Platform

Implementing a trading strategy

● Goals:o Write a simple algorithm to backtest our strategyo Compare the Wall Street estimates versus Estimize

estimates in generating alphao Get a sneak-peak into the Quantopian Research

Platform

Page 30: Crowdsource Earnings Predictions and the Quantopian Research Platform

The strategy

● PEAD - Post Earnings Announcement Drifto “The tendency for a stock’s cumulative abnormal

returns to drift in the direction of an earnings surprise”

● Logic:o If there is a positive surprise (actual > estimate)

Buy and hold for 1 day o If there is a negative surprise (actual < estimate)

Sell and hold for 1 day

Page 31: Crowdsource Earnings Predictions and the Quantopian Research Platform

Implementation

● Same format as Quantopian IDE:

Page 32: Crowdsource Earnings Predictions and the Quantopian Research Platform

Steps

1. Create a universe of stocks from our dataa. Only where N >= 20

2. Setup our `initialize` and `handle_data` methods

3. Run the algorithm4. Optimize our parameters and choose the

best one

Page 33: Crowdsource Earnings Predictions and the Quantopian Research Platform

Creating a universe

Page 34: Crowdsource Earnings Predictions and the Quantopian Research Platform

Initialize

Page 35: Crowdsource Earnings Predictions and the Quantopian Research Platform

Handle_data: Main Logic

Page 36: Crowdsource Earnings Predictions and the Quantopian Research Platform

Run the algorithm

Page 37: Crowdsource Earnings Predictions and the Quantopian Research Platform

Optimize parameters: Brute Force

● Our parameters: ● Run a for loop over these parameters:

o ● Redefined initialize with new params:

Page 38: Crowdsource Earnings Predictions and the Quantopian Research Platform

Results

Page 39: Crowdsource Earnings Predictions and the Quantopian Research Platform

Results

● The strategy that held a position for 3 days using the Estimize estimates had cumulative returns of 4.97% from 10/12 - 1/14

● Try:o Long onlyo Short onlyo Longer holding periods

Page 40: Crowdsource Earnings Predictions and the Quantopian Research Platform

Final thoughts and Summary

● There are more efficient ways to optimizeo Gradient descent, Walk forward optimization,

Genetic algorithms● Easier plotting tools exist (Seaborn)● Crowdsourced estimate data can be

interesting new sources of alpha● Parameter optimization/research is possible

in the Quantopian Research Platform

Page 41: Crowdsource Earnings Predictions and the Quantopian Research Platform

Questions and Notes● Email us at [email protected]

o Ask us about the iPython notebook these slides were based off!

● Visit us at Quantopiano www.quantopian.com

● Estimize Whitepaper: http://com.estimize.public.s3.amazonaws.com/papers/Estimize%20Whitepaper%20Executive%20Summary.pdf

● Deutsche Bank Paper: http://blog.estimize.com/post/80676086439/deutsche-bank-quant-research-estimize-more-timely-and