Zhongwu Huang

New corporate bond-level machine learning panel data with 341 predictors live on Open Source Bond Asset Pricing, OSBAP (https://lnkd.in/g_qDXCU4). Predictors are cross-sectionally ranked and re-scaled to the -1:1 interval. Merge this data with corporate bond return data on OSBAP and start forming factors with PyBondLab (https://lnkd.in/gaZeBU8d) or get some compute power and start data mining 😉. Example code for generating some factors from the 341 predictors can be found here: https://lnkd.in/guqs8CAa

104

3 Comments

The introduction of T 1 necessitated that all ecosystem participants – market makers, authorized participants, custodians and exchanges – collaborate to streamline operations. It was expected those innovations and adaptations would ultimately narrow spreads and maintain the competitiveness, efficiency and liquidity of ETFs. Read our latest T 1 article to learn more: http://ms.spr.ly/6045lSq1F #T1

1

☢️ I know some folks who follow my research are quite skeptical of data coming from the conference board, however I’m personally a big fan and the below chart represents a good example of why. The data is calling for a materially higher unemployment rate ahead based on the latest consumer confidence labor differential component reading. I‘ve been very clear about my position on labor market deterioration from the lagged impact of rate hikes, so I am biased with the chart’s implication. But what makes me partial to it is the historic relationship going back over 50 years and the consistency of results throughout numerous distinct macro backdrops. In each cycle we can clearly observe the labor differential component leading the broader US unemployment rate into recession. Unfortunately I’m quite confident history isn’t far away from repeating and markets should anticipate a higher unemployment rate ahead. Time will tell. Sources: Bloomberg, The Conference Board #Research #Economy #Markets #Macro #LaborMarket

27

3 Comments

PSA - Quarto is great for writing more than just Python/R reports - It's awesome for many types of documents. I found it really simple to set up a two-column layout with different languages in each column! The Typst (https://lnkd.in/eeiaZCej) rendering is also much nicer than LaTeX, and even supported non-Latin characters out of the box! If you've never used Quarto, but you use Jupyter notebooks, there are easy ways to take your notebooks and make them render prettily in HTML/PDF in Quarto (https://lnkd.in/echgUyiC) . (For those in the niche intersection of my interests – I am excited because I used it to create a source sheet 📄 for my Shavuot shiur last week ⛰ 🌾 ) Posit PBC Typst #quarto

45

1 Comment

While exploring Tableau Public, I stumbled upon a remarkable dashboard by Faith Rotich that really caught my attention. Here are three things I absolutely love about her work: 1️⃣ Symbolic Titles for Clear Navigation: Faith uses symbols in her titles to indicate the type of filter action on each visualization. This innovative approach makes it crystal clear how to interact with the dashboard—no guessing needed, just straightforward and efficient navigation. 2️⃣ Elegant Card Design with Rounded Corners: As a big fan of card layouts myself, I appreciate her use of a background image to create rounded corners. It's a subtle detail that adds extra flair and elevates the overall presentation. Sometimes, it's the small touches that make the biggest difference. 3️⃣ Intuitive Barbell Chart: The barbell chart she included provides an easy-to-understand view of the year-over-year profit changes by category. It's not just about displaying data but making it accessible and simple to grasp. Kudos to you, Faith Rotich! You've truly set the bar high. Keep pushing the boundaries of data visualization! #DataVisualization #Tableau #DataViz #DashboardDesign

70

1 Comment