R squared values

>>8607251
in the time you have been going back and forth you could have just calculated it by hand, you lazy faggot.

>>8607251
That looks very very wrong.

>>8607267
That's what machines are for, faggot. Until we get a machine that can write this thesis for me and create original thoughts, it can do my math for me.

>>8607273
In what way? There's no correlation between the independent and the dependent variable, therefore the R2 should be very low, right?

>>8607306
"Calculated by hand" as in write the fucking formula in R or Python and see for yourself what the R^2 value is.

And honestly, you should already have the data in a reasonable place like R and not fucking excel. You have a shit advisor if they allow that.

Low R2 values, especially values near zero, indicate that the model you've fitted is garbage and doesn't explain variation in data.

In this case, it means variation in pKH has no relation to variation in Equilibrium rejection.

Your advisor doesn't understand R2 values if they think it should be high. They're aren't just a measure of fit of a line to set of data, they're the result of applying a linear model to your data.

>>8607333
That's exactly what I've been saying! And in this case it's exactly what we want to see--for this range, we had expected that pKH would have no impact on equilibrium rejection.

>>8607251
It means that the data is explained better by its average value than by the linear regression.

>>8607251
gib raw data

>>8607883
k

12.19246497 -0.197273477
8.621602099 0.999732394
8.627087997 0.968863634
10.44611697 0.96966794
9.966576245 0.994183069
8.360513511 0.999643427
11.47755577 0.996342603
13.8827287 0.997151241
8.319664487 0.999936264
9.04769199 0.999794959
8.442492798 0.998327899
11.75696195 1
8.337242168 1
8.195860568 0.935675677
18.3990271 0.989191611
12.01954211 1
13.23284413 1
10.7878124 1
8.66756154 0.99782502

>>8608019
are you familiar with how r^2 is calculated?

>>8608094
Yes, although I usually just allow excel or R to do it for me.

Excel calculates R^2 as a measure of correlation between the dependent and independent variable--that is to say, how much the independent variable influences the dependent variable. r^2 is a measure of linearity.

>>8609751
For a simple single variate regression they're equivalent