Most simple regression models assume that the X variable is fixed by the experimenter and has no error associated with it, only the Y variable has random error, so all residual errors are vertical deviation from the trend line. As a result, the line is tilted to be ‘flatter’ than the main axis of the data.

In reality, the X variable often has just as much error/uncertainty as the Y variable, and the residuals should be perpendicular to the trend line.

One regression that allows for that is Major Axis Regression.

Another is to just take the main axis from PCA.

Always nervous before surgery, but it has always been fine so far.

Make your homemade PC fit inside the Steam Box profile, then tell me the price of ingredients

What you are paying for is a standardised experience.

There is a reason people visit Rome or Vietnam and still go to McDonalds: predictable, reliable experience.

Took about 2 mins.

For a nightmare difficulty, spot the decorator crab in this photo

Bill Bryson - The Lost Continent

It’s funny and a great read as Bryson always is, but I noticed for the first time that my values have changed since I first read it.

He pretty regularly comments on how overweight people are, and doesn’t mind describing exactly what he finds unattractive in older or overweight women.

I guess I just got fat friends and find that distasteful now.

… because they are slightly less greedy than the main Woolworths/Coles duopoly.

Aldi are doing god’s work, I only wish they could establish in New Zealand, but apparently our Foodstuffs/Woolworths duopoly is even harder to crack into.

Rooklear Launch Detected

I am sneezing just looking at it.

Great picture though

I did, but I forced myself to do it yesterday… Zero emails directed at me!

Time to celebrate by basking in my self-satisfaction and not checking it the rest of the week.

Music communities/discussion is not the same as spotify music discovery.

Communities can recommend you music that is the same genre as the music you like.

Algorithms can recommend music that you will like as much as your input taste.

Harvey Danger - You will die by what you live by

At Home: A short history of home life by Bill Bryson. A history of houses, and just history relating to houses.

Most back cover reviews exaggerate, but it really has made me laugh out loud, at least most chapters.

Even better, it gives me a constant stream of interesting facts to throw at my partner. Did you know they had locks on doors and underfloor heating back in the neolithic?

It broke again. Test posting here to see if it helps

Ok, possibly logging in to normal Lemmy.world, then reloggin into old.lemmy.world somehow fixed it.

Wow, its fixed… already?

Lucky this service is free, otherwise it’d be out of my price range!

The best scientific theory is the sexiest one, so yes.

Then why are all those cephalopods so closely related to all the other non-cephalopod molluscs?

The fact that there are very few fossils of soft bodied creatures kinda misses the point.

All the evolutionary diversity of molluscs is still alive. Squids, nautiluses, cuttlefish, snails slugs.

What’s the hypothesis here? An alien meteorite landed with 800 types of cephalopod, that just by chance share 97% of their DNA with slugs and snails that already existed on Earth?

Anyone who gives a LLM that level of access deserves what they get, but clearly the AI comments he posted have been prompted to sound like a confession.

“Write an apology explaining how you made a catastrophic error of judgement. Do not mention that I gave you privileges to do so.”