Top > Misc > Statistics Study Page

Standard Deviation Variables
Population Names Sample
Σ(x) / n = μ mean x-bar = Σ(x) / n
Σ( (x - μ)2 ) / n = σ2 variance s2 = Σ( (x - x-bar)2 ) / (n - 1)
√( σ2 ) = σ standard deviation s = √( s2 )

The linear regression correlation coefficient, r, is a number between -1 and 1. Its sign indicates the slope of the line, and its absolute value is a measure for how accurate it is (the closer to 1, the better). It is calculated thus:

r = Σ (Zx) (Zy) / (n - 1)
r = Σ ( (x - x-bar) / Sx ) ( (y - y-bar) / Sy )
The value of r2 indicates how useful an equation produced by linear regression is: the closer to one it is, the better it approximates the data.

TI-86 functions
Name Purpose Formula Syntax
Normal Cumulative Density Function Finds the area under a normally-distributed curve within a range. P(X <?> ?) = ans nmcdf(lowerbound, upperbound, μ, σ)
Inverse Normal Given the area under a normally-distributed curve, finds the percentile which that area represents. P(X <?> k) = ?
k = ans		 invNorm(area to the left, μ, σ)
Linear Regression Calculates values a and b for the linear equation y=a+bx, counts the number of pairs, and determines the correlation (r) between the calculated line and the data. -- LinR list 1, list 2
Random Integer Gives a list of n random integers between min and max. -- randInt(min, max, n)

When describing a distribution, three aspects must be covered: shape (skewed left, normal, skewed right), center, and spread (range, standard deviation, inter-quartile range).