#correlation

Did you calculate a #correlation coefficient? or a covariance? e.g., #MachineLearning feature selection, PCA? Cool, but did you check for outliers first? I shared my interactive #Python @matplotlib demonstration of the impact of a single outlier on the Pearson's and Spearman's correlation coefficients on #GitHub @ ∀. #DataScience

THE POWER OF CORRELATION ⬇️🫪 We are getting a better payout with a FLEX SLIP which is what we want anyways 😂 25x 6 man UNDERDOG CORRELATED SLIP 😤🔥 @OddsJam @EVBettors

How different are two-way fixed effects from modern difference-in-differences estimations? An AI-assisted review of 1,147 staggered adoption specifications finds estimates are highly correlated and sign flips are rare, from @instrumenthull

The book's edge grows with every leg you add to a parlay. Correlated legs flip that. When one hits, the next one is more likely to. Week 9 same game parlays, built on correlation:

"Over the past month, the 10y has become much more correlated to the Nasdaq than oil. This is a more risky regime...If the Fed decides that it must slow AI to calm inflation, then there is no easy ‘supply side’ end to the tightening cycle." @WarrenPies

Ever calculated a #correlation coefficient? Do you know what it looks like? Only under the assumption of bivariate #Gaussian does a correlation coefficient provide a complete description of your data, and it looks like this... ...linear and homoscedastic! See for yourself with my shared interactive #Python @matplotlib dashboard on #GitHub @ ∀. #DataScience

Thursday, November 5th, was a day of extremes: •US long/short Hedge Fund net leverage declined by 5% day over day, the 5th largest single-day decline since 2010. Net leverage is now in the bottom decile since 2010.^ •Long/short fund active gross flow since June has amounted to a -10.7% reduction relative to their gross market value at the end of June, the most in at least 10 years.^ •Hedge Funds are now underweight AI stocks* relative to market cap weights, but AI is still a very large percent of absolute exposures at 32%.^ •Levered ETFs had the largest amount to sell on data going back to 2007, bringing $57bn of US equity supply. •Retail selling of US equities in the first 90 minutes, based on MS QDS estimates using public data, was the second largest going back to 2016. •Equities are now starting to trade in a more correlated fashion, often signaling that risk reduction is broadening out. ^Sourced from MS Prime Brokerage. *AI stocks defined as the constituents of the MSXXAI Index, made up of AI tech beneficiaries, AI power names, and key hyperscalers

You do #MachineLearning? Do you calculate #correlation coefficients? How about covariance or PCA?, etc. OK, but beware! How do #outliers impact your results? ... totally! Check out my interactive #Python @matplotlib #Dashboard on #GitHub @ ∀. #DataScience

📊 Correlation is not causation—and not every trend is significant. More study hours may seem linked to higher scores, but statistical testing matters. Here, r = 0.234, which is not significant at the 5% level, so we fail to reject H₀. #Statistics #DataScience #Correlation #STEM

#LeadStoryOnET | #GDP #correlation gone for a toss, not sure how to read the 'real' numbers, says top industry figure

Smoking and drinking make your partner less happy too: - The Harvard Happiness Project found chronic smoking and chronic alcohol use are strongly anti-correlated with happiness - Partners and family members of chronic smokers and drinkers report lower happiness themselves, especially if they don't smoke or drink - Alcohol beyond 2 drinks per week is detrimental to health, a drink being a beer, a glass of wine or a cocktail @hubermanlab

Are we in a recession, or are people buying less beef because beef prices doubled? @costco #recession #correlation

Discover the key difference between #Correlation and #Regression that many analysts miss, explained by John V. Kane! 📊🔍 Read more: #DataScience #Statistics #DataAnalysis #DataAnalytics #rstats

Goldman (Wilson): I’d been arguing that this week (hyperscaler results) and next (mid-terms) would mark peak event risk for the market, and that clearing these 2 weeks would likely allow for a year-end rally, probably led by the biggest stocks in the index. I’d also argued that the AI sentiment pendulum had swung too far since end-Q2 and that substantive opportunities were beginning to emerge again. Well … ouch. Its been a pretty bruising couple of weeks navigating that view; for every step forward there’s been twice as many back. I don’t like the fact these notes have become so dominated by the AI debate, there’s so much else going on in the world... However, the reality is that the largest driver in macro markets today continues to be the AI investment cycle, and the resolution (either way) of newsflow in the coming weeks is still likely best expressed via many of the micro themes & trades in equity markets. As that AI-myopia and rotational pivot has amplified in markets, investors have been left to navigate unprecedented breakdowns in stock correlations: Apple is now negatively correlated to the Nasdaq for the first time ever.

How to NOT be confused by #Correlation: Why a coefficient of 0 does not mean ‘no relationship’ #DataScience

#Correlation & covariance are frequently applied in #DataAnalytics & #MachineLearning. Yet, they are quite sensitive to outliers. I made a interactive demonstration in #Python with @matplotlib to help my students see the power of a single #outlier @ ∀.

We mapped funding raised vs founder citations across 32 neo-labs Their founding teams have amassed 8.35M citations and have raised nearly $26B Discovery Loop leads with 1.59M founder citations, followed by SSI (915K) and AMI Labs (752K) The two are correlated, but a huge citation count certainly isn’t a prerequisite to be a neo-lab (see Standard Intelligence and Flapping Airplanes)

Correlation doesn't matter when yield farming alt/ETH or any crypto/crypto pairs. Why? Most crypto assets are correlated anyways. What we need to look at is relative strength. Divergence loss happens when the relative strength of two assets changes. Not correlation.

#Correlation does not imply #causation! Just because two things correlate does not mean that one causes the other! 📈📊 Correlations (or associations) between two things can be caused by a third factor that affects both of them.

When was the last time we saw Rus and NAZ nicely correlated on the intraday swings? Once again, profit taking on Rus while buying in NAZ. (MSFT ticked above ATH). Don't THINK.