Lecture_Statistics_Spring_2013b

# Iii probability distribution and confidence interval

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: he effect of degrees of freedom on the Student's t-distribution and how it approaches a standard normal distribution as N → ∞. 0.4 Probability 0.3 z Distribution µ = 0, σ = 1 t Distributions ν = 30 ν=8 ν=3 0.2 0.1 0.0 -4 -2 0 t or z 2 4 See Figure 3 III. Probability Distributions & Confidence Intervals N observations (N-1 degrees of freedom) with the known experimental variance and mean a 100(1-α)% confidence interval is given as: ȹ s ȹ x ± t α / 2, υ = N −1 ȹ ȹ ȹ ȹ ȹ N Ⱥ ( ) Meaning of confidence interval: A 95% confidence interval means that 95% of all confidence intervals constructed according to the above equation will cover the true mean. III. Probability Distribution and Confidence Interval Weighted Average ȹ m ȹ ȹ m n i ȹ ȹ m ȹ x = ȹ ∑ n i x i ȹ N = ȹ ∑ x i ȹ = ȹ ∑ x i p(x i )ȹ ȹ ȹ ȹ N ȹ ȹ ȹ ȹ i =1 Ⱥ ȹ i =1 Ⱥ ȹ i =1 Ⱥ m: the number of unique observations and Ni: the number of times a given xi value is observed. *Weighted averages are important in chemistry Examples: atomic weights on a periodic table are weighted fro isotopes. IV. Propagation of Error How to determine the standard deviation of a multivariable function Q = f (X1, X2, ..., XN) comprised of N variables, X1, X2, X3, ….etc. 1) Take the total differential of this function: ȹ ∂Q ȹ ȹ ∂Q ȹ ȹ ∂Q ȹ ȹ ȹ ȹ ȹ dQ = ȹ dX1 + ȹ dX 2 + … + ȹ dX N...
View Full Document

## This note was uploaded on 01/26/2014 for the course CHEM 3625 taught by Professor Mrjohnson during the Spring '08 term at Virginia Tech.

Ask a homework question - tutors are online