Does anyone else on /sci/ keep track of data in their day-to-day life? Pic related, graph of my daily electricity consumption vs the average temperature. Makes a nice example of quadratic regression.
>>8742258
>°F
ugh
>>8742258
>quadratic regression
>not using generalized least squares for function spaces
your stats are shit use better modeling
>>8742258
No, but it sounds neat. Like fun little hobby.
Then again, this seemingly harmless hobby sounds a lot like those people who keep live journals and have stress balls and other "neat" gadgets to help sort out their life but then end up even more stressed because know they have a daily live journal schedule to follow.
I'll do this if I ever stop having a real job, so that this doesn't interfere with my daily life.
What's the best way to measure time spent at activities? For instance in a day I could do : x hours sleep, y hours math, z hours shit post, etc; so how would I plot all those variables on a day-to-day basis?
>>8742258
keep track of my weight, thinking about adding a field for poops and food
how do you collect your data? I assume its automated in some way
>>8742629
I download the data from my electricity provider. And I get the weather data from weather underground.
>>8742271
Yes, that's the temperature of preference in my country. Not saying I approve of it.
>>8742280
>The estimates of both OLS and GLS should be close to one another, if not numerically then in the size of the “impact”. If they are not, then it would most likely indicate that you have a function form misspecification(s), of that you have left out variables.
https://stats.stackexchange.com/questions/155031/how-to-determine-if-gls-improves-on-ols
Please go have sexual intercourse with yourself or the farm animal of your choice :^)
>>8742271
This
>not using kelvin
>>8742715
>defending your shit models instead of just using a superior algorithm
classic brainlet
>>8742280
>Using GLS instead of MLE
Your stats are shit, use better modeling.
>>8742271
>Bitching about english units yet he actually types out noises people make when they can't think of words to say
>>8744599
Shit model? It's a simple bivariate relationship for a basic graph, not a statistics thesis. There's no reason to use GLS in this situation.
• timed bike-ride to school for a year
• amazed at small Standard Deviation
• modelled with Gamma distribution
• show my Statistics Professor
• extra credit for Special Project
>>8745147
post it faggoit
>>8742258
That is cool, OP. How did you make it?
>>8746156
Gnumeric. It's my favorite little spreadsheet program.
>>8746132
didn't save, sorry faggot