Quant Screening Backtesting: Turnaround Stocks

Posted on May 4, 2017

Introduction: What are turnaround stocks?

Turnaround investing is the process of looking for investment opportunities in down-and-out companies that are poised to experience a financial recovery.


In this way, turnaround plays aka deep-value stories are among the most profitable strategies among value investors; mind you, investors willing to invest in these stocks could fall victim of two main risks:

  1. Cost of opportunity: the cost opportunity is crucial as more often than not wounded companies take more than expected to heal and reflect such improvement in their stock price.
  2. Downside risk: entry point miscalculation is one of the most frequent mistakes in unexperienced investors. An impatient investor recklessly looking for deep-value picks might commit a grievous mistake when entering too early into a troubled company and, eventually, suffer significant losses that undermine absolute returns and the risk-reward contribution for the portfolio.

Both risks will be discussed in this post as we answered several questions related to a turnaround screening approach and its backtesting analysis.

Turnaround Stocks: Research Literature findings

Scholar and practitioner literature is substantial but not as vast as for other topics or so-called valuation or trading anomalies. A summary of the most ground-breaking works is shown in the table below. Multiple authors have delved deeply into several dimensions of the turnaround effect such as the management role, recovery stages, leading or financial ratios recovery indicators:


Screening Criteria: Dividend Omissions as Key Indicator

Buland and Subramanian (2008 and 2012) identified dividend omission events as a key point for successful turnaround stories. Dividend omissions -  cutting dividend completely - act as a kick off point in a change corporate behaviour and can be easily measured quantitatively. In fact, a “Quantamental” investment approach combining quantitative and fundamental reasoning is helpful to reduce a preliminary human bias to shortlist popular stories, summarize numerically tons of past research insights and optimize the amount of time allocated to identify turnaround candidates.

Bearing in mind the aforementioned, a turnaround quantitative screening criteria  has been backtested considering more than 15 self-help factors such as dividend omissions or management changes. A sample of the five factors with the closest link to the turnaround strategy rationale is shown below:

  • Depressed profitability levels: financial statement items at extremely low levels.
  • Dividend cut: recent negative dividend per share growth with stocks with dividend omission (100% dividend cut) scoring higher.
  • Change in Management: CEO, COO and CFO replacements before dividend cut announcement.
  • Operating momentum recovering: recent improvement on margins on a sequential and annual basis.
  • Safe short term credit status: improvement in key credit metrics such as Interest coverage or Net Debt-to-EBITDA.

Backtesting Analysis:

Bloomberg and Python have been used for this quantitative screening backtesting analysis while utilizing R and Shiny web development tools for data analysis purposes. The screened universe of companies was comprised of more than 9,000 stocks from the US, UK and European developed markets for the period 2003-2015 with only 215 shortlisted using six-month screening rebalancing periods. Total returns (capital appreciation plus dividends) and dividends are assumed to be reinvested when received. Survivorship bias is taken into account as there are several stocks filing for Chapter 11 and 7 that yield a completely loss of invested capital in the analysis.

The objective of the following analysis is to obtain a  preliminary set of insightful conclusions answering key questions that can help both smart-beta investors eager to discover strategies to be applied using a top-down approach as well as more bottom-up biased stock pickers seeking new screening methods to optimize their candidate shortlisting process:

  • What type of companies are more prone to meet the screening criteria?
  • Which is the best investment horizon range for a turnaround investor?
  • What are the risks a turnaround investor faces when implementing the strategy?
  • Is the turnaround strategy providing significant alpha?
  • Are there any specific sectors or cyclicality levels where the strategy is more profitable?
  • What is the best allocation from a risk-reward and win ratio standpoint?
  • What’s the Bayesian probability of ending a trade with positive P&L when the first six months the position is underwater?

What type of companies are more prone to meet the screening criteria?

According to our screening results, turnaround plays are not occurring uniformly across the board: turnaround plays are not occurring uniformly across the board. More than half of the screened companies below to sectors with significant business cyclicality bias (sensitiveness to economic cycle) whereas less than one quarter of the sample belongs to more defensive. Hence, an investor narrowing his portfolio to a turnaround strategy is to bear a huge cyclical bias.

Cyclicality Pie

Which is the best investment horizon range for a turnaround investor?

Total returns for six different investment horizons are considered spanning from six months to three years. On the one hand, turnaround stocks cumulative return profile is mainly positively correlated as the investment time horizon increases. On the other hand, annualized returns shrink as the holding period increases.  That said, Long term turnaround investors can pocket +15% CAGR for holding periods above 1.5 years, yet short term turnaround investors able to identify repeatedly new turnaround investments to replace old stories can maximize CAGR to levels above +30% CAGR.


What are the risks a turnaround investor faces when implementing the strategy?

Cost of opportunity risk is important as mentioned in the earlier question for those investors willing to deliver annualized returns above 15%. However, a paramount risk that overshadows any other is bankruptcy as it occurs even when a stock meets some sanity checks from our criteria. In addition, the distribution of CAGR returns for the different investment horizons is highlighting significant negative skew (average below median) and negative kurtosis across all the horizons. Because of this “tail risk” feature, popular risk measures such as standard deviation are not recommended to be used when assessing a turnaround strategy risk profile.


Is the turnaround strategy providing significant alpha?

Conclusions about total cumulative alpha and CAGR alpha are similar to the ones drawn from our analysis on absolute returns: cumulative alpha increases with time; yet CAGR alpha (annualised alpha) also exhibits a diminishing return profile as time passes. Overall, long term turnaround investors might outperform local benchmarks by approximately 25% (holding period between 1.5 years and 3 years) whereas a short term investor with one year horizon can reap an exceptional CAGR alpha near +50%. The strategy also passes the net return acid test: the high-performance results and low rebalancing assumption (six months) yield significant performance results even when assuming 400 bps annual charge for trading fees and bid-offer spread costs.

Are there any specific sectors or cyclicality levels where the strategy is more profitable?

An initial conclusion is that short term traders should bias their trading books towards cyclical turnaround stories as displayed in the couple of boxplot charts below. Nevertheless, long term investors can afford some more defensiveness in their portfolios. Mind you, this analysis ignores other important key metrics beyond returns related to risk and risk-adjusted returns that are answered in the next question.


What is the best allocation from a risk-reward and win ratio standpoint?

When comparing hit ratio (percentage of winning trades) compared to risk-reward (average positive return over average negative return) for both short term and long term holding periods, we confirm that short term traders should bias their books towards more cyclical companies where long term investors might prefer some more neutral-defensive bias. The first chart below shows how stocks with cyclicality score 'very high','high' and 'low' score the highest in terms of both risk-reward and hit ratio for short term investment period (six months):


That said, cyclical turnaround plays for long term horizons seem to have a much poorer "hit ratio vs risk-reward" profile in the long term than defensive stocks. The plot below supports this observation with only very defensive stocks ( low cyclicality score) populating the top-right region:


What’s the Bayesian probability of ending a trade with positive P&L when the first six months the position is underwater?

Overoptimism is one of the most dangerous behavioural biases in finance. Investors are reluctant to sell losing trades hoping a low probability turnaround is coming. Bad luck, macro conditions and a myriad of other excuses are set forth by them to refrain from closing their position. Overoptimism comes hand-to-hand with other biases such as cognitive dissonance (self-denial when material evidence against the initial thesis exists) and confirmation bias (empathy gap aka falling in love with the stock).

A good way to test if it makes sense to exit an unsuccessful trade is to analyse what happens with our turnaround strategy during the first 6 months and compare to future results. Bayesian probability theory is very helpful here: Bayesian inference is a method of statistical inference in which Bayes' theorem is used to update the probability for a hypothesis as more evidence or information becomes available:


The questions to be answered is: What’s the probability to end up a trade with positive return when the first six months the trade are negative? Applying the Bayes theorem to our dataset is straightforward as it shown in the table below:

  1. Negative total return (first six months): probability of ending with a positive return for each investment horizon assuming the stock total return is negative during the first six months.
  2. Positive total return (first six months): probability of ending with a positive return for each investment horizon assuming the stock total return is positive during the first six months.
  3. Positive start – negative start: difference between 2 and 1 that aids to visualize whether or not the initial condition is really a game-changer in terms of winning probability.


The main takeaway is that bad starts tend to have a low probability of success: less than 20% in all the holding periods except for the 3-year horizon. This is particularly underwhelming when compared to the probability of success of good starts: more than 30% difference in terms of probability between good starts and bad starts probabilities.

However, when considering only negative bad starts as those where a stock is in negative territory is unfair. For instance, the stock can have a bad start due to broad market conditions or simply being down only by 2%, which is not very significant. The table below filters out this effects considering stocks falling below 15% a bad start (first row) and those falling less than 15% or in positive territory as a good start (second row):


The lesson to learn here is that a stop-loss level like 15% for turnaround stocks during the first six months of trading boosts the success rate of a turnaround strategy. Hence, turnaround investors should bear in mind implementing a disciplined stop-loss policy in order to enhance their performance metrics.

Click here to check code in Github

About Author

Carlos Salas Najera

Carlos is a passionate individual for investments and technology with a long/short equity analysis and portfolio management experience approach that combines his fundamental, quantitative and data science skills in order to deliver superior returns. His core strength lies...
View all posts by Carlos Salas Najera >

Related Articles

Leave a Comment

No comments found.

View Posts by Categories

Our Recent Popular Posts

View Posts by Tags

#python #trainwithnycdsa 2019 2020 Revenue 3-points agriculture air quality airbnb airline alcohol Alex Baransky algorithm alumni Alumni Interview Alumni Reviews Alumni Spotlight alumni story Alumnus ames dataset ames housing dataset apartment rent API Application artist aws bank loans beautiful soup Best Bootcamp Best Data Science 2019 Best Data Science Bootcamp Best Data Science Bootcamp 2020 Best Ranked Big Data Book Launch Book-Signing bootcamp Bootcamp Alumni Bootcamp Prep boston safety Bundles cake recipe California Cancer Research capstone car price Career Career Day citibike classic cars classpass clustering Coding Course Demo Course Report covid 19 credit credit card crime frequency crops D3.js data data analysis Data Analyst data analytics data for tripadvisor reviews data science Data Science Academy Data Science Bootcamp Data science jobs Data Science Reviews Data Scientist Data Scientist Jobs data visualization database Deep Learning Demo Day Discount disney dplyr drug data e-commerce economy employee employee burnout employer networking environment feature engineering Finance Financial Data Science fitness studio Flask flight delay gbm Get Hired ggplot2 googleVis H20 Hadoop hallmark holiday movie happiness healthcare frauds higgs boson Hiring hiring partner events Hiring Partners hotels housing housing data housing predictions housing price hy-vee Income Industry Experts Injuries Instructor Blog Instructor Interview insurance italki Job Job Placement Jobs Jon Krohn JP Morgan Chase Kaggle Kickstarter las vegas airport lasso regression Lead Data Scienctist Lead Data Scientist leaflet league linear regression Logistic Regression machine learning Maps market matplotlib Medical Research Meet the team meetup methal health miami beach movie music Napoli NBA netflix Networking neural network Neural networks New Courses NHL nlp NYC NYC Data Science nyc data science academy NYC Open Data nyc property NYCDSA NYCDSA Alumni Online Online Bootcamp Online Training Open Data painter pandas Part-time performance phoenix pollutants Portfolio Development precision measurement prediction Prework Programming public safety PwC python Python Data Analysis python machine learning python scrapy python web scraping python webscraping Python Workshop R R Data Analysis R language R Programming R Shiny r studio R Visualization R Workshop R-bloggers random forest Ranking recommendation recommendation system regression Remote remote data science bootcamp Scrapy scrapy visualization seaborn seafood type Selenium sentiment analysis sentiment classification Shiny Shiny Dashboard Spark Special Special Summer Sports statistics streaming Student Interview Student Showcase SVM Switchup Tableau teachers team team performance TensorFlow Testimonial tf-idf Top Data Science Bootcamp Top manufacturing companies Transfers tweets twitter videos visualization wallstreet wallstreetbets web scraping Weekend Course What to expect whiskey whiskeyadvocate wildfire word cloud word2vec XGBoost yelp youtube trending ZORI