it might be harder for me to get the good advice that I need from my doctor than it would be if we were consulting regression , a predictive model, some other source of information.My question for you is, as we head into a world where we're going to have more access to predictive models, more artificial intelligence, more thorough use of machine
three things together to help us understand meaning now I think of it this way until my NBA students this all the time that when we are doing regressions or we're doing quantitative research the regression for instance orthe statistical uh analysis that's the research instrument but when we're doing qualitative work we are studying culture
culture divide is so strong here that even when someone tries to explain the culture divide he ends up looking a regressions and logistic regressions we did a lot of that then we did some research on how to redo our analyses
blue State book He's author on several other books basing on data analysis teaching statistics a bag of tricks and also a book about using data analysis using regression multi-level hierarchical models um he's a prolific researcher on a lot of a lot of topics including rationality of voting and the probability that your vote will be decisive um I voted in Florida in 2000
OK, so that's all techniques we talked about for cheap restaurants. We ran a little regression on price.
And were chance findings that we perform replication, they go away. But the possibility of regression to the mean is there for other situations. What's the difference between those two?
Their languages are also being destroyed. There's also a regression .
Their languages are also being destroyed. That's a regression that goes back 800 years.
And the tokens are worth $1 to each of them but $2 to their partner. these are standardized regression coefficients.
Erik Brynjolfsson is a friend at MIT. I ran my regressions past him just to make sure I had done it right and he said, "Yeah, it all looks good and damn I wish my regressions looked like that, too." Jon Koomey: but doubling time for computations per kilowatt hour, 1.6 years or so for all computers, 1.5 years for PCs and slightly
As men-- we see in research-- acquire more experience about how investing works, they tend to reduce their level of risk, and the opposite This is known as a regression to the mean, right?
in German society. So we are going back to some kind of class society. But it is not a regression to former societies. It's based on a contradiction and this simultaneity of progress and regression .
And the answer is about 25%. This is an r-squared form of regression , for those of you who remember, . So in other words, these company accounts here-- very uninformative about the set of assets which are really powering a company.
And it turns out there were four recurring patterns. I had the research team do 99 regression analyses. And they finally said, enough.
was on Google. And this is public data aggregate from across the world of what people are actually looking up for "I always." that they're facing other than linear regression and ratio.
But curiosity, as Flexner writes, it goes back-- it's an outstanding characteristic of modern thinking. So we get this kind of regression to the mean, and a lot of boring research is being done.
years ago, from 2050. If we projected the same linear regression trend to 2050. So, today if we're 2016, you can see where we are, roughly 11,000 per capita.
And then they looked through multiple regression analyses, they were able to determine over time who was still standing and breathing
And this is a CAT scan showing regression of those tumors in this patient's lung.
These guys did a whole bunch of regressions .
add as controls or as explanatory factors in a regression model, like a confounder.
What he actually does is a little multivariate regression .
And this is what we find as we start running these regression models.
I'm here to help reclaim the lost kingdom faded in regression .
So this all goes together to this big regression model.
So that's what I mean by automatic regression .
So this is the results of the first regression analysis that I did.
Then we're going to do a linear regression across every pixel in the map.
It could be, to some extent. And some of these diminishing effects could be regression to the mean for some effects that are genuine but inflated when we first discover them. And I think I'll show some examples about that.
Their languages are also being destroyed. Well, that's regression . And they're going on in parallel.
And the tokens are worth $1 to each of them but $2 to their partner. And if you look at it-- again these are regression coefficients-- if you tapped your hand in time with this person, yes, you felt more similar to them.
And you're not supposed to scrutinize this regression analysis.
That's mostly the dependent variable in the regressions that I'll tell you about in a minute.
We have a lot of people who can do regressions .
our our quality and our repeatability and not have so many regressions and all
And also, you get a regression coefficient that tells you which ones are the most important and increase receptiveness the most.
And in blue, if we plot a regression line through those points, we see that it has a negative slope.
And again, if we put a regression line through that part, we get a correlation of negative 0.8.
and add them up-- the algorithm always wins. By algorithm here, I mean 1950s-era regression equation. All the more so when the machine learning is more sophisticated than just adding up a bunch of variables.
But it is not a regression to former societies. It's based on a contradiction and this simultaneity of progress and regression . We have more horizontal equality, but at the same time, more vertical inequality, which means--
And then all the information that we collected goes into one big regression model.
All right, what's automatic regression ?
This is exactly the kind of intermediate technology you could stop the further regression of your society and start pulling yourself back up by your own bootstraps.
in the general social survey and we were able to replicate our regression analysis with that.
Their languages are also being destroyed. There's progress, there's regression .
What we do on it is we're able to run this multivariate regression analysis where we can separate out-- Using your smart meter data and the local weather data stream, the
After that, we ran logistic regressions .
can do the statistical analysis, and run the regressions and see what the factors were that led to the success.
For example, we're going to run this logistic regression model, and we're going to judge the best option by what's the most significant parameter
did with low birthweight, or if you add a collider as a control variable in a regression model, you are going to induce a biased view of what