Forming Earning Season Themes: Moving away from memory management (mini project)

Most agree Earnings season is an important component for investors in building an overall Market Risk picture for the short/medium term outlook.

But practically forming that view, for me, is the ART of consuming as much good information as possible before memory overload. By better managing the information (Bloomberg + Python) I hope to better understand what the overall trend is, outliers and focus areas.

Practically its me Exploring Bloomberg's Python interface with a focus on understanding trends in the Earnings season . A mini project if you would to build connections within markets and data science, expanding current skill set in different areas of finance and in this case Python.

I thought this is interesting enough to share, especially given:

1. Its a really easy start, given with 3 lines of code you can produce a great dashboard to setoff the journey. Good time now too as near the tail end of the season (473 of 491 stocks within US 500 large Cap )

Code above produces the dashboard which is a great place to start forming a EPS thematic view based on the numbers. I'll highlight a few takeaways I found through the data.

The scatterplot above shows EPS Beat vs.EPS Surprise, but a number of differentanalysis can be viewed from the default setting including:

• EPS Surprise or Revenue Surprise vs:
• EPS broker estimate dispersion, EPS revisions, Momentum, Quality, Value or sales growth

The dashboard is also interactive enabling the user to click a specific bubble and get underling summary, like the below for Alphabet Inc.

2. Besides the dashboard example above, there are a number of Equity examples in the spotlight webinars which make some of the below observations possible.

Webinar details here:

Bquant Spotlight: Exploring Equity Broker Estimates in Bquant

Bquant Basics: Equities

Here are a few key takeaways the experts at Bloomberg have found within the latest EPS data.

Personally I cannot replicate all (yet), but will continu as Ithink understanding how to geth these insights alone is already great, but opens doors to further insights not yet known possible.

Price response shows lack of Enthusiasm for companies beating expectations

Nearly 90% of companies beat expectations on the season, the strongest rate in our records back to 1992. 2.1% of reports topped revenue-growth estimates, second-best since 3Q09, and 17.9% missed, the lowest in our records since 2009. Stock-price reactions to 1Q beats were muted, whilst the response to misses was extreme - suggesting the market largely expected big upside surprises.

Expectations for the future are even higher - Guidance supports recovery

Analysts made notable upward adjustments to 2021 earnings expectations amid much stronger-than-anticipated 1Q earnings reports, and much of the recovery is expected to stretch into 2022.

NLP analysis within transcripts

• particularly focusing on inflation this season

Mentions of inflation in earnings-call transcripts in 1Q outpaced any period since 1Q16. Judging by transcripts, the strongest direct impact from price increases so far is in the household and personal products and capital goods industries

• particularly focusing on supply chain Constraints

Supply chain mentions shot up in conference calls this season, matching references to the word last recorded when tariff announcements first surprised financial markets in early 2018, and suggesting a growing amount of anxiety over product sourcing

Understanding the main 5 factor group themes better.

Recent analysis of median EPS beats of our five main factor groups would be of interest for me to underhand how exactly this is done in BQNT

Though this earnings season was broadly strong, analyzing median EPS surprises for Q1 portfolios in each of our five main U.S. equity factor groups reveals dispersion in earnings results. Momentum and High value stocks reported strong earnings, while high-quality and low-volatility groups had more modest EPS beats this season.

Food for thought.

What you think. interesting?

Please share if you have dived in here and can help share in the progress.