Degrees of Freedom (Statistics)

[Anonymous No.9062438]

Will someone explain to me what exactly degrees of freedom means in stats?

[Anonymous No.9062516]

Accounting for variance in variables and some shit

[Anonymous No.9062871]

>>9062516
yes

[Anonymous No.9062878]

>>9062516
this isnt exactly an enlightening response, this much is generally gathered already

[Anonymous No.9062960]

>>9062438

Straight from Wikipedia and actually surprisingly helpful:

>In general, the degrees of freedom of an estimate of a parameter are equal to the number of independent scores that go into the estimate minus the number of parameters used as intermediate steps in the estimation of the parameter itself

>(e.g. the sample variance has N − 1 degrees of freedom, since it is computed from N random scores minus the only 1 parameter estimated as intermediate step, which is the sample mean).

Did you even try before making your shitpost, anon?

[Anonymous No.9063504]

>>9062960
Th-thanks.

[Anonymous No.9063523]

>>9062438
It's kinda like how many different ways a particular data point can change. So think of it like moving on a two dimensional grid. You measure the absolute movement in any direction but there is a contribution from both the x and y direction for any movement made, even if it is just zero. So when you gather data in the x direction for this absolute movement you have to account for the fact that your subject could have moved in the y direction as well throwing off the correlation between the x data set and the absolute movement data set. In science the goal is to reduce the number of affecting variables or fix them into known ranges and values so only the parameter of interest changes, ie. putting walls on either side of the subject before taking absolute movement data so it can't move in the y direction.

At least that's how I think about it.

[Anonymous No.9063551]

>>9062960
worse definition ever, you can find the variance without calculating the mean by using the moment-generating function.